---
_id: '13315'
abstract:
- lang: eng
text: How do statistical dependencies in measurement noise influence high-dimensional
inference? To answer this, we study the paradigmatic spiked matrix model of principal
components analysis (PCA), where a rank-one matrix is corrupted by additive noise.
We go beyond the usual independence assumption on the noise entries, by drawing
the noise from a low-order polynomial orthogonal matrix ensemble. The resulting
noise correlations make the setting relevant for applications but analytically
challenging. We provide characterization of the Bayes optimal limits of inference
in this model. If the spike is rotation invariant, we show that standard spectral
PCA is optimal. However, for more general priors, both PCA and the existing approximate
message-passing algorithm (AMP) fall short of achieving the information-theoretic
limits, which we compute using the replica method from statistical physics. We
thus propose an AMP, inspired by the theory of adaptive Thouless–Anderson–Palmer
equations, which is empirically observed to saturate the conjectured theoretical
limit. This AMP comes with a rigorous state evolution analysis tracking its performance.
Although we focus on specific noise distributions, our methodology can be generalized
to a wide class of trace matrix ensembles at the cost of more involved expressions.
Finally, despite the seemingly strong assumption of rotation-invariant noise,
our theory empirically predicts algorithmic performance on real data, pointing
at strong universality properties.
acknowledgement: J.B. was funded by the European Union (ERC, CHORAL, project number
101039794). Views and opinions expressed are however those of the author(s) only
and do not necessarily reflect those of the European Union or the European Research
Council. Neither the European Union nor the granting authority can be held responsible
for them. M.M. was supported by the 2019 Lopez-Loreta Prize. We would like to thank
the reviewers for the insightful comments and, in particular, for suggesting the
BAMP-inspired denoisers leading to AMP-AP.
article_number: e2302028120
article_processing_charge: Yes (in subscription journal)
article_type: original
author:
- first_name: Jean
full_name: Barbier, Jean
last_name: Barbier
- first_name: Francesco
full_name: Camilli, Francesco
last_name: Camilli
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
- first_name: Manuel
full_name: Sáenz, Manuel
last_name: Sáenz
citation:
ama: Barbier J, Camilli F, Mondelli M, Sáenz M. Fundamental limits in structured
principal component analysis and how to reach them. Proceedings of the National
Academy of Sciences of the United States of America. 2023;120(30). doi:10.1073/pnas.2302028120
apa: Barbier, J., Camilli, F., Mondelli, M., & Sáenz, M. (2023). Fundamental
limits in structured principal component analysis and how to reach them. Proceedings
of the National Academy of Sciences of the United States of America. National
Academy of Sciences. https://doi.org/10.1073/pnas.2302028120
chicago: Barbier, Jean, Francesco Camilli, Marco Mondelli, and Manuel Sáenz. “Fundamental
Limits in Structured Principal Component Analysis and How to Reach Them.” Proceedings
of the National Academy of Sciences of the United States of America. National
Academy of Sciences, 2023. https://doi.org/10.1073/pnas.2302028120.
ieee: J. Barbier, F. Camilli, M. Mondelli, and M. Sáenz, “Fundamental limits in
structured principal component analysis and how to reach them,” Proceedings
of the National Academy of Sciences of the United States of America, vol.
120, no. 30. National Academy of Sciences, 2023.
ista: Barbier J, Camilli F, Mondelli M, Sáenz M. 2023. Fundamental limits in structured
principal component analysis and how to reach them. Proceedings of the National
Academy of Sciences of the United States of America. 120(30), e2302028120.
mla: Barbier, Jean, et al. “Fundamental Limits in Structured Principal Component
Analysis and How to Reach Them.” Proceedings of the National Academy of Sciences
of the United States of America, vol. 120, no. 30, e2302028120, National Academy
of Sciences, 2023, doi:10.1073/pnas.2302028120.
short: J. Barbier, F. Camilli, M. Mondelli, M. Sáenz, Proceedings of the National
Academy of Sciences of the United States of America 120 (2023).
date_created: 2023-07-30T22:01:02Z
date_published: 2023-07-25T00:00:00Z
date_updated: 2023-10-17T11:44:55Z
day: '25'
ddc:
- '000'
department:
- _id: MaMo
doi: 10.1073/pnas.2302028120
external_id:
pmid:
- '37463204'
file:
- access_level: open_access
checksum: 1fc06228afdb3aa80cf8e7766bcf9dc5
content_type: application/pdf
creator: dernst
date_created: 2023-07-31T07:30:48Z
date_updated: 2023-07-31T07:30:48Z
file_id: '13323'
file_name: 2023_PNAS_Barbier.pdf
file_size: 995933
relation: main_file
success: 1
file_date_updated: 2023-07-31T07:30:48Z
has_accepted_license: '1'
intvolume: ' 120'
issue: '30'
language:
- iso: eng
month: '07'
oa: 1
oa_version: Published Version
pmid: 1
project:
- _id: 059876FA-7A3F-11EA-A408-12923DDC885E
name: Prix Lopez-Loretta 2019 - Marco Mondelli
publication: Proceedings of the National Academy of Sciences of the United States
of America
publication_identifier:
eissn:
- 1091-6490
publication_status: published
publisher: National Academy of Sciences
quality_controlled: '1'
related_material:
link:
- relation: software
url: https://github.com/fcamilli95/Structured-PCA-
scopus_import: '1'
status: public
title: Fundamental limits in structured principal component analysis and how to reach
them
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 120
year: '2023'
...
---
_id: '14037'
abstract:
- lang: eng
text: 'Traditionally, nuclear spin is not considered to affect biological processes.
Recently, this has changed as isotopic fractionation that deviates from classical
mass dependence was reported both in vitro and in vivo. In these cases, the isotopic
effect correlates with the nuclear magnetic spin. Here, we show nuclear spin effects
using stable oxygen isotopes (16O, 17O, and 18O) in two separate setups: an artificial
dioxygen production system and biological aquaporin channels in cells. We observe
that oxygen dynamics in chiral environments (in particular its transport) depend
on nuclear spin, suggesting future applications for controlled isotope separation
to be used, for instance, in NMR. To demonstrate the mechanism behind our findings,
we formulate theoretical models based on a nuclear-spin-enhanced switch between
electronic spin states. Accounting for the role of nuclear spin in biology can
provide insights into the role of quantum effects in living systems and help inspire
the development of future biotechnology solutions.'
acknowledgement: N.M.-S. acknowledges the support of the Ministry of Energy, Israel,
as part of the scholarship program for graduate students in the fields of energy.
M.L. acknowledges support by the European Research Council (ERC) Starting Grant
No. 801770 (ANGULON). Y.P. acknowledges the support of the Ministry of Innovation,
Science and Technology, Israel Grant No. 1001593872. Y.P acknowledges the support
of the BSF-NSF 094 Grant No. 2022503.
article_number: e2300828120
article_processing_charge: Yes (in subscription journal)
article_type: original
author:
- first_name: Ofek
full_name: Vardi, Ofek
last_name: Vardi
- first_name: Naama
full_name: Maroudas-Sklare, Naama
last_name: Maroudas-Sklare
- first_name: Yuval
full_name: Kolodny, Yuval
last_name: Kolodny
- first_name: Artem
full_name: Volosniev, Artem
id: 37D278BC-F248-11E8-B48F-1D18A9856A87
last_name: Volosniev
orcid: 0000-0003-0393-5525
- first_name: Amijai
full_name: Saragovi, Amijai
last_name: Saragovi
- first_name: Nir
full_name: Galili, Nir
last_name: Galili
- first_name: Stav
full_name: Ferrera, Stav
last_name: Ferrera
- first_name: Areg
full_name: Ghazaryan, Areg
id: 4AF46FD6-F248-11E8-B48F-1D18A9856A87
last_name: Ghazaryan
orcid: 0000-0001-9666-3543
- first_name: Nir
full_name: Yuran, Nir
last_name: Yuran
- first_name: Hagit P.
full_name: Affek, Hagit P.
last_name: Affek
- first_name: Boaz
full_name: Luz, Boaz
last_name: Luz
- first_name: Yonaton
full_name: Goldsmith, Yonaton
last_name: Goldsmith
- first_name: Nir
full_name: Keren, Nir
last_name: Keren
- first_name: Shira
full_name: Yochelis, Shira
last_name: Yochelis
- first_name: Itay
full_name: Halevy, Itay
last_name: Halevy
- first_name: Mikhail
full_name: Lemeshko, Mikhail
id: 37CB05FA-F248-11E8-B48F-1D18A9856A87
last_name: Lemeshko
orcid: 0000-0002-6990-7802
- first_name: Yossi
full_name: Paltiel, Yossi
last_name: Paltiel
citation:
ama: Vardi O, Maroudas-Sklare N, Kolodny Y, et al. Nuclear spin effects in biological
processes. Proceedings of the National Academy of Sciences of the United States
of America. 2023;120(32). doi:10.1073/pnas.2300828120
apa: Vardi, O., Maroudas-Sklare, N., Kolodny, Y., Volosniev, A., Saragovi, A., Galili,
N., … Paltiel, Y. (2023). Nuclear spin effects in biological processes. Proceedings
of the National Academy of Sciences of the United States of America. National
Academy of Sciences. https://doi.org/10.1073/pnas.2300828120
chicago: Vardi, Ofek, Naama Maroudas-Sklare, Yuval Kolodny, Artem Volosniev, Amijai
Saragovi, Nir Galili, Stav Ferrera, et al. “Nuclear Spin Effects in Biological
Processes.” Proceedings of the National Academy of Sciences of the United States
of America. National Academy of Sciences, 2023. https://doi.org/10.1073/pnas.2300828120.
ieee: O. Vardi et al., “Nuclear spin effects in biological processes,” Proceedings
of the National Academy of Sciences of the United States of America, vol.
120, no. 32. National Academy of Sciences, 2023.
ista: Vardi O, Maroudas-Sklare N, Kolodny Y, Volosniev A, Saragovi A, Galili N,
Ferrera S, Ghazaryan A, Yuran N, Affek HP, Luz B, Goldsmith Y, Keren N, Yochelis
S, Halevy I, Lemeshko M, Paltiel Y. 2023. Nuclear spin effects in biological processes.
Proceedings of the National Academy of Sciences of the United States of America.
120(32), e2300828120.
mla: Vardi, Ofek, et al. “Nuclear Spin Effects in Biological Processes.” Proceedings
of the National Academy of Sciences of the United States of America, vol.
120, no. 32, e2300828120, National Academy of Sciences, 2023, doi:10.1073/pnas.2300828120.
short: O. Vardi, N. Maroudas-Sklare, Y. Kolodny, A. Volosniev, A. Saragovi, N. Galili,
S. Ferrera, A. Ghazaryan, N. Yuran, H.P. Affek, B. Luz, Y. Goldsmith, N. Keren,
S. Yochelis, I. Halevy, M. Lemeshko, Y. Paltiel, Proceedings of the National Academy
of Sciences of the United States of America 120 (2023).
date_created: 2023-08-13T22:01:12Z
date_published: 2023-07-31T00:00:00Z
date_updated: 2023-10-17T11:45:25Z
day: '31'
ddc:
- '530'
department:
- _id: MiLe
doi: 10.1073/pnas.2300828120
ec_funded: 1
external_id:
pmid:
- '37523549'
file:
- access_level: open_access
checksum: a5ed64788a5acef9b9a300a26fa5a177
content_type: application/pdf
creator: dernst
date_created: 2023-08-14T07:43:45Z
date_updated: 2023-08-14T07:43:45Z
file_id: '14047'
file_name: 2023_PNAS_Vardi.pdf
file_size: 1003092
relation: main_file
success: 1
file_date_updated: 2023-08-14T07:43:45Z
has_accepted_license: '1'
intvolume: ' 120'
issue: '32'
language:
- iso: eng
month: '07'
oa: 1
oa_version: Published Version
pmid: 1
project:
- _id: 2688CF98-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '801770'
name: 'Angulon: physics and applications of a new quasiparticle'
publication: Proceedings of the National Academy of Sciences of the United States
of America
publication_identifier:
eissn:
- 1091-6490
publication_status: published
publisher: National Academy of Sciences
quality_controlled: '1'
scopus_import: '1'
status: public
title: Nuclear spin effects in biological processes
tmp:
image: /images/cc_by_nc_nd.png
legal_code_url: https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode
name: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
(CC BY-NC-ND 4.0)
short: CC BY-NC-ND (4.0)
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 120
year: '2023'
...
---
_id: '12683'
abstract:
- lang: eng
text: We study the eigenvalue trajectories of a time dependent matrix Gt=H+itvv∗
for t≥0, where H is an N×N Hermitian random matrix and v is a unit vector. In
particular, we establish that with high probability, an outlier can be distinguished
at all times t>1+N−1/3+ϵ, for any ϵ>0. The study of this natural process combines
elements of Hermitian and non-Hermitian analysis, and illustrates some aspects
of the intrinsic instability of (even weakly) non-Hermitian matrices.
acknowledgement: G. Dubach gratefully acknowledges funding from the European Union’s
Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie
Grant Agreement No. 754411. L. Erdős is supported by ERC Advanced Grant “RMTBeyond”
No. 101020331.
article_processing_charge: No
article_type: original
author:
- first_name: Guillaume
full_name: Dubach, Guillaume
id: D5C6A458-10C4-11EA-ABF4-A4B43DDC885E
last_name: Dubach
orcid: 0000-0001-6892-8137
- first_name: László
full_name: Erdös, László
id: 4DBD5372-F248-11E8-B48F-1D18A9856A87
last_name: Erdös
orcid: 0000-0001-5366-9603
citation:
ama: Dubach G, Erdös L. Dynamics of a rank-one perturbation of a Hermitian matrix.
Electronic Communications in Probability. 2023;28:1-13. doi:10.1214/23-ECP516
apa: Dubach, G., & Erdös, L. (2023). Dynamics of a rank-one perturbation of
a Hermitian matrix. Electronic Communications in Probability. Institute
of Mathematical Statistics. https://doi.org/10.1214/23-ECP516
chicago: Dubach, Guillaume, and László Erdös. “Dynamics of a Rank-One Perturbation
of a Hermitian Matrix.” Electronic Communications in Probability. Institute
of Mathematical Statistics, 2023. https://doi.org/10.1214/23-ECP516.
ieee: G. Dubach and L. Erdös, “Dynamics of a rank-one perturbation of a Hermitian
matrix,” Electronic Communications in Probability, vol. 28. Institute of
Mathematical Statistics, pp. 1–13, 2023.
ista: Dubach G, Erdös L. 2023. Dynamics of a rank-one perturbation of a Hermitian
matrix. Electronic Communications in Probability. 28, 1–13.
mla: Dubach, Guillaume, and László Erdös. “Dynamics of a Rank-One Perturbation of
a Hermitian Matrix.” Electronic Communications in Probability, vol. 28,
Institute of Mathematical Statistics, 2023, pp. 1–13, doi:10.1214/23-ECP516.
short: G. Dubach, L. Erdös, Electronic Communications in Probability 28 (2023) 1–13.
date_created: 2023-02-26T23:01:01Z
date_published: 2023-02-08T00:00:00Z
date_updated: 2023-10-17T12:48:10Z
day: '08'
ddc:
- '510'
department:
- _id: LaEr
doi: 10.1214/23-ECP516
ec_funded: 1
external_id:
arxiv:
- '2108.13694'
isi:
- '000950650200005'
file:
- access_level: open_access
checksum: a1c6f0a3e33688fd71309c86a9aad86e
content_type: application/pdf
creator: dernst
date_created: 2023-02-27T09:43:27Z
date_updated: 2023-02-27T09:43:27Z
file_id: '12692'
file_name: 2023_ElectCommProbability_Dubach.pdf
file_size: 479105
relation: main_file
success: 1
file_date_updated: 2023-02-27T09:43:27Z
has_accepted_license: '1'
intvolume: ' 28'
isi: 1
language:
- iso: eng
month: '02'
oa: 1
oa_version: Published Version
page: 1-13
project:
- _id: 260C2330-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '754411'
name: ISTplus - Postdoctoral Fellowships
- _id: 62796744-2b32-11ec-9570-940b20777f1d
call_identifier: H2020
grant_number: '101020331'
name: Random matrices beyond Wigner-Dyson-Mehta
publication: Electronic Communications in Probability
publication_identifier:
eissn:
- 1083-589X
publication_status: published
publisher: Institute of Mathematical Statistics
quality_controlled: '1'
scopus_import: '1'
status: public
title: Dynamics of a rank-one perturbation of a Hermitian matrix
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 28
year: '2023'
...
---
_id: '12761'
abstract:
- lang: eng
text: "We consider the fluctuations of regular functions f of a Wigner matrix W
viewed as an entire matrix f (W). Going beyond the well-studied tracial mode,
Trf (W), which is equivalent to the customary linear statistics of eigenvalues,
we show that Trf (W)A is asymptotically normal for any nontrivial bounded deterministic
matrix A. We identify three different and asymptotically independent modes of
this fluctuation, corresponding to the tracial part, the traceless diagonal part
and the off-diagonal part of f (W) in the entire mesoscopic regime, where we find
that the off-diagonal modes fluctuate on a much smaller scale than the tracial
mode. As a main motivation to study CLT in such generality on small mesoscopic
scales, we determine\r\nthe fluctuations in the eigenstate thermalization hypothesis
(Phys. Rev. A 43 (1991) 2046–2049), that is, prove that the eigenfunction overlaps
with any deterministic matrix are asymptotically Gaussian after a small spectral
averaging. Finally, in the macroscopic regime our result also generalizes (Zh.
Mat. Fiz. Anal. Geom. 9 (2013) 536–581, 611, 615) to complex W and to all crossover
ensembles in between. The main technical inputs are the recent\r\nmultiresolvent
local laws with traceless deterministic matrices from the companion paper (Comm.
Math. Phys. 388 (2021) 1005–1048)."
acknowledgement: The second author is partially funded by the ERC Advanced Grant “RMTBEYOND”
No. 101020331. The third author is supported by Dr. Max Rössler, the Walter Haefner
Foundation and the ETH Zürich Foundation.
article_processing_charge: No
article_type: original
author:
- first_name: Giorgio
full_name: Cipolloni, Giorgio
id: 42198EFA-F248-11E8-B48F-1D18A9856A87
last_name: Cipolloni
orcid: 0000-0002-4901-7992
- first_name: László
full_name: Erdös, László
id: 4DBD5372-F248-11E8-B48F-1D18A9856A87
last_name: Erdös
orcid: 0000-0001-5366-9603
- first_name: Dominik J
full_name: Schröder, Dominik J
id: 408ED176-F248-11E8-B48F-1D18A9856A87
last_name: Schröder
orcid: 0000-0002-2904-1856
citation:
ama: Cipolloni G, Erdös L, Schröder DJ. Functional central limit theorems for Wigner
matrices. Annals of Applied Probability. 2023;33(1):447-489. doi:10.1214/22-AAP1820
apa: Cipolloni, G., Erdös, L., & Schröder, D. J. (2023). Functional central
limit theorems for Wigner matrices. Annals of Applied Probability. Institute
of Mathematical Statistics. https://doi.org/10.1214/22-AAP1820
chicago: Cipolloni, Giorgio, László Erdös, and Dominik J Schröder. “Functional Central
Limit Theorems for Wigner Matrices.” Annals of Applied Probability. Institute
of Mathematical Statistics, 2023. https://doi.org/10.1214/22-AAP1820.
ieee: G. Cipolloni, L. Erdös, and D. J. Schröder, “Functional central limit theorems
for Wigner matrices,” Annals of Applied Probability, vol. 33, no. 1. Institute
of Mathematical Statistics, pp. 447–489, 2023.
ista: Cipolloni G, Erdös L, Schröder DJ. 2023. Functional central limit theorems
for Wigner matrices. Annals of Applied Probability. 33(1), 447–489.
mla: Cipolloni, Giorgio, et al. “Functional Central Limit Theorems for Wigner Matrices.”
Annals of Applied Probability, vol. 33, no. 1, Institute of Mathematical
Statistics, 2023, pp. 447–89, doi:10.1214/22-AAP1820.
short: G. Cipolloni, L. Erdös, D.J. Schröder, Annals of Applied Probability 33 (2023)
447–489.
date_created: 2023-03-26T22:01:08Z
date_published: 2023-02-01T00:00:00Z
date_updated: 2023-10-17T12:48:52Z
day: '01'
department:
- _id: LaEr
doi: 10.1214/22-AAP1820
ec_funded: 1
external_id:
arxiv:
- '2012.13218'
isi:
- '000946432400015'
intvolume: ' 33'
isi: 1
issue: '1'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/2012.13218
month: '02'
oa: 1
oa_version: Preprint
page: 447-489
project:
- _id: 62796744-2b32-11ec-9570-940b20777f1d
call_identifier: H2020
grant_number: '101020331'
name: Random matrices beyond Wigner-Dyson-Mehta
publication: Annals of Applied Probability
publication_identifier:
issn:
- 1050-5164
publication_status: published
publisher: Institute of Mathematical Statistics
quality_controlled: '1'
scopus_import: '1'
status: public
title: Functional central limit theorems for Wigner matrices
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 33
year: '2023'
...
---
_id: '8682'
abstract:
- lang: eng
text: It is known that the Brauer--Manin obstruction to the Hasse principle is vacuous
for smooth Fano hypersurfaces of dimension at least 3 over any number field. Moreover,
for such varieties it follows from a general conjecture of Colliot-Thélène that
the Brauer--Manin obstruction to the Hasse principle should be the only one, so
that the Hasse principle is expected to hold. Working over the field of rational
numbers and ordering Fano hypersurfaces of fixed degree and dimension by height,
we prove that almost every such hypersurface satisfies the Hasse principle provided
that the dimension is at least 3. This proves a conjecture of Poonen and Voloch
in every case except for cubic surfaces.
article_processing_charge: No
article_type: original
author:
- first_name: Timothy D
full_name: Browning, Timothy D
id: 35827D50-F248-11E8-B48F-1D18A9856A87
last_name: Browning
orcid: 0000-0002-8314-0177
- first_name: Pierre Le
full_name: Boudec, Pierre Le
last_name: Boudec
- first_name: Will
full_name: Sawin, Will
last_name: Sawin
citation:
ama: Browning TD, Boudec PL, Sawin W. The Hasse principle for random Fano hypersurfaces.
Annals of Mathematics. 2023;197(3):1115-1203. doi:10.4007/annals.2023.197.3.3
apa: Browning, T. D., Boudec, P. L., & Sawin, W. (2023). The Hasse principle
for random Fano hypersurfaces. Annals of Mathematics. Princeton University.
https://doi.org/10.4007/annals.2023.197.3.3
chicago: Browning, Timothy D, Pierre Le Boudec, and Will Sawin. “The Hasse Principle
for Random Fano Hypersurfaces.” Annals of Mathematics. Princeton University,
2023. https://doi.org/10.4007/annals.2023.197.3.3.
ieee: T. D. Browning, P. L. Boudec, and W. Sawin, “The Hasse principle for random
Fano hypersurfaces,” Annals of Mathematics, vol. 197, no. 3. Princeton
University, pp. 1115–1203, 2023.
ista: Browning TD, Boudec PL, Sawin W. 2023. The Hasse principle for random Fano
hypersurfaces. Annals of Mathematics. 197(3), 1115–1203.
mla: Browning, Timothy D., et al. “The Hasse Principle for Random Fano Hypersurfaces.”
Annals of Mathematics, vol. 197, no. 3, Princeton University, 2023, pp.
1115–203, doi:10.4007/annals.2023.197.3.3.
short: T.D. Browning, P.L. Boudec, W. Sawin, Annals of Mathematics 197 (2023) 1115–1203.
date_created: 2020-10-19T14:28:50Z
date_published: 2023-05-01T00:00:00Z
date_updated: 2023-10-17T12:47:43Z
day: '01'
department:
- _id: TiBr
doi: 10.4007/annals.2023.197.3.3
external_id:
arxiv:
- '2006.02356'
isi:
- '000966611000003'
intvolume: ' 197'
isi: 1
issue: '3'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://arxiv.org/abs/2006.02356
month: '05'
oa: 1
oa_version: Preprint
page: 1115-1203
publication: Annals of Mathematics
publication_identifier:
issn:
- 0003-486X
publication_status: published
publisher: Princeton University
quality_controlled: '1'
related_material:
link:
- description: News on IST Homepage
relation: press_release
url: https://ist.ac.at/en/news/when-is-necessary-sufficient/
status: public
title: The Hasse principle for random Fano hypersurfaces
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 197
year: '2023'
...
---
_id: '12706'
abstract:
- lang: eng
text: Allometric settings of population dynamics models are appealing due to their
parsimonious nature and broad utility when studying system level effects. Here,
we parameterise the size-scaled Rosenzweig-MacArthur differential equations to
eliminate prey-mass dependency, facilitating an in depth analytic study of the
equations which incorporates scaling parameters’ contributions to coexistence.
We define the functional response term to match empirical findings, and examine
situations where metabolic theory derivations and observation diverge. The dynamical
properties of the Rosenzweig-MacArthur system, encompassing the distribution of
size-abundance equilibria, the scaling of period and amplitude of population cycling,
and relationships between predator and prey abundances, are consistent with empirical
observation. Our parameterisation is an accurate minimal model across 15+ orders
of mass magnitude.
acknowledgement: "This research was supported by an Australian Government Research
Training Program\r\n(RTP) Scholarship to JCM (https://www.dese.gov.au), and LB is
supported by the Centre de\r\nrecherche sur le vieillissement Fellowship Program.
The funders had no role in study design, data collection and analysis, decision
to publish, or preparation of the manuscript."
article_processing_charge: No
article_type: original
author:
- first_name: Jody C.
full_name: Mckerral, Jody C.
last_name: Mckerral
- first_name: Maria
full_name: Kleshnina, Maria
id: 4E21749C-F248-11E8-B48F-1D18A9856A87
last_name: Kleshnina
- first_name: Vladimir
full_name: Ejov, Vladimir
last_name: Ejov
- first_name: Louise
full_name: Bartle, Louise
last_name: Bartle
- first_name: James G.
full_name: Mitchell, James G.
last_name: Mitchell
- first_name: Jerzy A.
full_name: Filar, Jerzy A.
last_name: Filar
citation:
ama: Mckerral JC, Kleshnina M, Ejov V, Bartle L, Mitchell JG, Filar JA. Empirical
parameterisation and dynamical analysis of the allometric Rosenzweig-MacArthur
equations. PLoS One. 2023;18(2):e0279838. doi:10.1371/journal.pone.0279838
apa: Mckerral, J. C., Kleshnina, M., Ejov, V., Bartle, L., Mitchell, J. G., &
Filar, J. A. (2023). Empirical parameterisation and dynamical analysis of the
allometric Rosenzweig-MacArthur equations. PLoS One. Public Library of
Science. https://doi.org/10.1371/journal.pone.0279838
chicago: Mckerral, Jody C., Maria Kleshnina, Vladimir Ejov, Louise Bartle, James
G. Mitchell, and Jerzy A. Filar. “Empirical Parameterisation and Dynamical Analysis
of the Allometric Rosenzweig-MacArthur Equations.” PLoS One. Public Library
of Science, 2023. https://doi.org/10.1371/journal.pone.0279838.
ieee: J. C. Mckerral, M. Kleshnina, V. Ejov, L. Bartle, J. G. Mitchell, and J. A.
Filar, “Empirical parameterisation and dynamical analysis of the allometric Rosenzweig-MacArthur
equations,” PLoS One, vol. 18, no. 2. Public Library of Science, p. e0279838,
2023.
ista: Mckerral JC, Kleshnina M, Ejov V, Bartle L, Mitchell JG, Filar JA. 2023. Empirical
parameterisation and dynamical analysis of the allometric Rosenzweig-MacArthur
equations. PLoS One. 18(2), e0279838.
mla: Mckerral, Jody C., et al. “Empirical Parameterisation and Dynamical Analysis
of the Allometric Rosenzweig-MacArthur Equations.” PLoS One, vol. 18, no.
2, Public Library of Science, 2023, p. e0279838, doi:10.1371/journal.pone.0279838.
short: J.C. Mckerral, M. Kleshnina, V. Ejov, L. Bartle, J.G. Mitchell, J.A. Filar,
PLoS One 18 (2023) e0279838.
date_created: 2023-03-05T23:01:05Z
date_published: 2023-02-27T00:00:00Z
date_updated: 2023-10-17T12:53:30Z
day: '27'
ddc:
- '000'
department:
- _id: KrCh
doi: 10.1371/journal.pone.0279838
external_id:
isi:
- '000996122900022'
pmid:
- '36848357'
file:
- access_level: open_access
checksum: 798ed5739a4117b03173e5d56e0534c9
content_type: application/pdf
creator: cchlebak
date_created: 2023-03-07T10:26:45Z
date_updated: 2023-03-07T10:26:45Z
file_id: '12712'
file_name: 2023_PLOSOne_Mckerral.pdf
file_size: 1257003
relation: main_file
success: 1
file_date_updated: 2023-03-07T10:26:45Z
has_accepted_license: '1'
intvolume: ' 18'
isi: 1
issue: '2'
language:
- iso: eng
month: '02'
oa: 1
oa_version: Published Version
page: e0279838
pmid: 1
publication: PLoS One
publication_identifier:
eissn:
- 1932-6203
publication_status: published
publisher: Public Library of Science
quality_controlled: '1'
scopus_import: '1'
status: public
title: Empirical parameterisation and dynamical analysis of the allometric Rosenzweig-MacArthur
equations
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 18
year: '2023'
...
---
_id: '13202'
abstract:
- lang: eng
text: Phosphatidylinositol-4,5-bisphosphate (PI(4,5)P2) plays an essential role
in neuronal activities through interaction with various proteins involved in signaling
at membranes. However, the distribution pattern of PI(4,5)P2 and the association
with these proteins on the neuronal cell membranes remain elusive. In this study,
we established a method for visualizing PI(4,5)P2 by SDS-digested freeze-fracture
replica labeling (SDS-FRL) to investigate the quantitative nanoscale distribution
of PI(4,5)P2 in cryo-fixed brain. We demonstrate that PI(4,5)P2 forms tiny clusters
with a mean size of ∼1000 nm2 rather than randomly distributed in cerebellar neuronal
membranes in male C57BL/6J mice. These clusters show preferential accumulation
in specific membrane compartments of different cell types, in particular, in Purkinje
cell (PC) spines and granule cell (GC) presynaptic active zones. Furthermore,
we revealed extensive association of PI(4,5)P2 with CaV2.1 and GIRK3 across different
membrane compartments, whereas its association with mGluR1α was compartment specific.
These results suggest that our SDS-FRL method provides valuable insights into
the physiological functions of PI(4,5)P2 in neurons.
acknowledged_ssus:
- _id: EM-Fac
acknowledgement: This work was supported by The Institute of Science and Technology
(IST) Austria, the European Union's Horizon 2020 Research and Innovation Program
under the Marie Skłodowska-Curie Grant Agreement No. 793482 (to K.E.) and by the
European Research Council (ERC) Grant Agreement No. 694539 (to R.S.). We thank Nicoleta
Condruz (IST Austria, Klosterneuburg, Austria) for technical assistance with sample
preparation, the Electron Microscopy Facility of IST Austria (Klosterneuburg, Austria)
for technical support with EM works, Natalia Baranova (University of Vienna, Vienna,
Austria) and Martin Loose (IST Austria, Klosterneuburg, Austria) for advice on liposome
preparation, and Yugo Fukazawa (University of Fukui, Fukui, Japan) for comments.
article_processing_charge: No
article_type: original
author:
- first_name: Kohgaku
full_name: Eguchi, Kohgaku
id: 2B7846DC-F248-11E8-B48F-1D18A9856A87
last_name: Eguchi
orcid: 0000-0002-6170-2546
- first_name: Elodie
full_name: Le Monnier, Elodie
id: 3B59276A-F248-11E8-B48F-1D18A9856A87
last_name: Le Monnier
- first_name: Ryuichi
full_name: Shigemoto, Ryuichi
id: 499F3ABC-F248-11E8-B48F-1D18A9856A87
last_name: Shigemoto
orcid: 0000-0001-8761-9444
citation:
ama: Eguchi K, Le Monnier E, Shigemoto R. Nanoscale phosphoinositide distribution
on cell membranes of mouse cerebellar neurons. The Journal of Neuroscience.
2023;43(23):4197-4216. doi:10.1523/JNEUROSCI.1514-22.2023
apa: Eguchi, K., Le Monnier, E., & Shigemoto, R. (2023). Nanoscale phosphoinositide
distribution on cell membranes of mouse cerebellar neurons. The Journal of
Neuroscience. Society for Neuroscience. https://doi.org/10.1523/JNEUROSCI.1514-22.2023
chicago: Eguchi, Kohgaku, Elodie Le Monnier, and Ryuichi Shigemoto. “Nanoscale Phosphoinositide
Distribution on Cell Membranes of Mouse Cerebellar Neurons.” The Journal of
Neuroscience. Society for Neuroscience, 2023. https://doi.org/10.1523/JNEUROSCI.1514-22.2023.
ieee: K. Eguchi, E. Le Monnier, and R. Shigemoto, “Nanoscale phosphoinositide distribution
on cell membranes of mouse cerebellar neurons,” The Journal of Neuroscience,
vol. 43, no. 23. Society for Neuroscience, pp. 4197–4216, 2023.
ista: Eguchi K, Le Monnier E, Shigemoto R. 2023. Nanoscale phosphoinositide distribution
on cell membranes of mouse cerebellar neurons. The Journal of Neuroscience. 43(23),
4197–4216.
mla: Eguchi, Kohgaku, et al. “Nanoscale Phosphoinositide Distribution on Cell Membranes
of Mouse Cerebellar Neurons.” The Journal of Neuroscience, vol. 43, no.
23, Society for Neuroscience, 2023, pp. 4197–216, doi:10.1523/JNEUROSCI.1514-22.2023.
short: K. Eguchi, E. Le Monnier, R. Shigemoto, The Journal of Neuroscience 43 (2023)
4197–4216.
date_created: 2023-07-09T22:01:12Z
date_published: 2023-06-07T00:00:00Z
date_updated: 2023-10-18T07:12:47Z
day: '07'
ddc:
- '570'
department:
- _id: RySh
doi: 10.1523/JNEUROSCI.1514-22.2023
ec_funded: 1
external_id:
isi:
- '001020132100005'
pmid:
- '37160366'
file:
- access_level: open_access
checksum: 70b2141870e0bf1c94fd343e18fdbc32
content_type: application/pdf
creator: alisjak
date_created: 2023-07-10T09:04:58Z
date_updated: 2023-07-10T09:04:58Z
file_id: '13205'
file_name: 2023_JN_Eguchi.pdf
file_size: 7794425
relation: main_file
success: 1
file_date_updated: 2023-07-10T09:04:58Z
has_accepted_license: '1'
intvolume: ' 43'
isi: 1
issue: '23'
language:
- iso: eng
month: '06'
oa: 1
oa_version: Published Version
page: 4197-4216
pmid: 1
project:
- _id: 2659CC84-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '793482'
name: 'Ultrastructural analysis of phosphoinositides in nerve terminals: distribution,
dynamics and physiological roles in synaptic transmission'
- _id: 25CA28EA-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '694539'
name: 'In situ analysis of single channel subunit composition in neurons: physiological
implication in synaptic plasticity and behaviour'
publication: The Journal of Neuroscience
publication_identifier:
eissn:
- 1529-2401
issn:
- 0270-6474
publication_status: published
publisher: Society for Neuroscience
quality_controlled: '1'
scopus_import: '1'
status: public
title: Nanoscale phosphoinositide distribution on cell membranes of mouse cerebellar
neurons
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 43
year: '2023'
...
---
_id: '12916'
abstract:
- lang: eng
text: "We apply a variant of the square-sieve to produce an upper bound for the
number of rational points of bounded height on a family of surfaces that admit
a fibration over P1 whose general fibre is a hyperelliptic curve. The implied
constant does not depend on the coefficients of the polynomial defining the surface.\r\n"
article_processing_charge: No
article_type: original
author:
- first_name: Dante
full_name: Bonolis, Dante
id: 6A459894-5FDD-11E9-AF35-BB24E6697425
last_name: Bonolis
- first_name: Timothy D
full_name: Browning, Timothy D
id: 35827D50-F248-11E8-B48F-1D18A9856A87
last_name: Browning
orcid: 0000-0002-8314-0177
citation:
ama: Bonolis D, Browning TD. Uniform bounds for rational points on hyperelliptic
fibrations. Annali della Scuola Normale Superiore di Pisa - Classe di Scienze.
2023;24(1):173-204. doi:10.2422/2036-2145.202010_018
apa: Bonolis, D., & Browning, T. D. (2023). Uniform bounds for rational points
on hyperelliptic fibrations. Annali Della Scuola Normale Superiore Di Pisa
- Classe Di Scienze. Scuola Normale Superiore - Edizioni della Normale. https://doi.org/10.2422/2036-2145.202010_018
chicago: Bonolis, Dante, and Timothy D Browning. “Uniform Bounds for Rational Points
on Hyperelliptic Fibrations.” Annali Della Scuola Normale Superiore Di Pisa
- Classe Di Scienze. Scuola Normale Superiore - Edizioni della Normale, 2023.
https://doi.org/10.2422/2036-2145.202010_018.
ieee: D. Bonolis and T. D. Browning, “Uniform bounds for rational points on hyperelliptic
fibrations,” Annali della Scuola Normale Superiore di Pisa - Classe di Scienze,
vol. 24, no. 1. Scuola Normale Superiore - Edizioni della Normale, pp. 173–204,
2023.
ista: Bonolis D, Browning TD. 2023. Uniform bounds for rational points on hyperelliptic
fibrations. Annali della Scuola Normale Superiore di Pisa - Classe di Scienze.
24(1), 173–204.
mla: Bonolis, Dante, and Timothy D. Browning. “Uniform Bounds for Rational Points
on Hyperelliptic Fibrations.” Annali Della Scuola Normale Superiore Di Pisa
- Classe Di Scienze, vol. 24, no. 1, Scuola Normale Superiore - Edizioni della
Normale, 2023, pp. 173–204, doi:10.2422/2036-2145.202010_018.
short: D. Bonolis, T.D. Browning, Annali Della Scuola Normale Superiore Di Pisa
- Classe Di Scienze 24 (2023) 173–204.
date_created: 2023-05-07T22:01:04Z
date_published: 2023-02-16T00:00:00Z
date_updated: 2023-10-18T06:54:30Z
day: '16'
department:
- _id: TiBr
doi: 10.2422/2036-2145.202010_018
external_id:
arxiv:
- '2007.14182'
intvolume: ' 24'
issue: '1'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://doi.org/10.48550/arXiv.2007.14182
month: '02'
oa: 1
oa_version: Preprint
page: 173-204
publication: Annali della Scuola Normale Superiore di Pisa - Classe di Scienze
publication_identifier:
eissn:
- 2036-2145
issn:
- 0391-173X
publication_status: published
publisher: Scuola Normale Superiore - Edizioni della Normale
quality_controlled: '1'
scopus_import: '1'
status: public
title: Uniform bounds for rational points on hyperelliptic fibrations
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 24
year: '2023'
...
---
_id: '14422'
abstract:
- lang: eng
text: "Animals exhibit a remarkable ability to learn and remember new behaviors,
skills, and associations throughout their lifetime. These capabilities are made
possible thanks to a variety of\r\nchanges in the brain throughout adulthood,
regrouped under the term \"plasticity\". Some cells\r\nin the brain —neurons—
and specifically changes in the connections between neurons, the\r\nsynapses,
were shown to be crucial for the formation, selection, and consolidation of memories\r\nfrom
past experiences. These ongoing changes of synapses across time are called synaptic\r\nplasticity.
Understanding how a myriad of biochemical processes operating at individual\r\nsynapses
can somehow work in concert to give rise to meaningful changes in behavior is
a\r\nfascinating problem and an active area of research.\r\nHowever, the experimental
search for the precise plasticity mechanisms at play in the brain\r\nis daunting,
as it is difficult to control and observe synapses during learning. Theoretical\r\napproaches
have thus been the default method to probe the plasticity-behavior connection.
Such\r\nstudies attempt to extract unifying principles across synapses and model
all observed synaptic\r\nchanges using plasticity rules: equations that govern
the evolution of synaptic strengths across\r\ntime in neuronal network models.
These rules can use many relevant quantities to determine\r\nthe magnitude of
synaptic changes, such as the precise timings of pre- and postsynaptic\r\naction
potentials, the recent neuronal activity levels, the state of neighboring synapses,
etc.\r\nHowever, analytical studies rely heavily on human intuition and are forced
to make simplifying\r\nassumptions about plasticity rules.\r\nIn this thesis,
we aim to assist and augment human intuition in this search for plasticity rules.\r\nWe
explore whether a numerical approach could automatically discover the plasticity
rules\r\nthat elicit desired behaviors in large networks of interconnected neurons.
This approach is\r\ndubbed meta-learning synaptic plasticity: learning plasticity
rules which themselves will make\r\nneuronal networks learn how to solve a desired
task. We first write all the potential plasticity\r\nmechanisms to consider using
a single expression with adjustable parameters. We then optimize\r\nthese plasticity
parameters using evolutionary strategies or Bayesian inference on tasks known\r\nto
involve synaptic plasticity, such as familiarity detection and network stabilization.\r\nWe
show that these automated approaches are powerful tools, able to complement established\r\nanalytical
methods. By comprehensively screening plasticity rules at all synapse types in\r\nrealistic,
spiking neuronal network models, we discover entire sets of degenerate plausible\r\nplasticity
rules that reliably elicit memory-related behaviors. Our approaches allow for
more\r\nrobust experimental predictions, by abstracting out the idiosyncrasies
of individual plasticity\r\nrules, and provide fresh insights on synaptic plasticity
in spiking network models.\r\n"
alternative_title:
- ISTA Thesis
article_processing_charge: No
author:
- first_name: Basile J
full_name: Confavreux, Basile J
id: C7610134-B532-11EA-BD9F-F5753DDC885E
last_name: Confavreux
citation:
ama: 'Confavreux BJ. Synapseek: Meta-learning synaptic plasticity rules. 2023. doi:10.15479/at:ista:14422'
apa: 'Confavreux, B. J. (2023). Synapseek: Meta-learning synaptic plasticity
rules. Institute of Science and Technology Austria. https://doi.org/10.15479/at:ista:14422'
chicago: 'Confavreux, Basile J. “Synapseek: Meta-Learning Synaptic Plasticity Rules.”
Institute of Science and Technology Austria, 2023. https://doi.org/10.15479/at:ista:14422.'
ieee: 'B. J. Confavreux, “Synapseek: Meta-learning synaptic plasticity rules,” Institute
of Science and Technology Austria, 2023.'
ista: 'Confavreux BJ. 2023. Synapseek: Meta-learning synaptic plasticity rules.
Institute of Science and Technology Austria.'
mla: 'Confavreux, Basile J. Synapseek: Meta-Learning Synaptic Plasticity Rules.
Institute of Science and Technology Austria, 2023, doi:10.15479/at:ista:14422.'
short: 'B.J. Confavreux, Synapseek: Meta-Learning Synaptic Plasticity Rules, Institute
of Science and Technology Austria, 2023.'
date_created: 2023-10-12T14:13:25Z
date_published: 2023-10-12T00:00:00Z
date_updated: 2023-10-18T09:20:56Z
day: '12'
ddc:
- '610'
degree_awarded: PhD
department:
- _id: GradSch
- _id: TiVo
doi: 10.15479/at:ista:14422
ec_funded: 1
file:
- access_level: closed
checksum: 7f636555eae7803323df287672fd13ed
content_type: application/pdf
creator: cchlebak
date_created: 2023-10-12T14:53:50Z
date_updated: 2023-10-12T14:54:52Z
embargo: 2024-10-12
embargo_to: open_access
file_id: '14424'
file_name: Confavreux_Thesis_2A.pdf
file_size: 30599717
relation: main_file
- access_level: closed
checksum: 725e85946db92290a4583a0de9779e1b
content_type: application/x-zip-compressed
creator: cchlebak
date_created: 2023-10-18T07:38:34Z
date_updated: 2023-10-18T07:56:08Z
file_id: '14440'
file_name: Confavreux Thesis.zip
file_size: 68406739
relation: source_file
file_date_updated: 2023-10-18T07:56:08Z
has_accepted_license: '1'
language:
- iso: eng
license: https://creativecommons.org/licenses/by-nc-sa/4.0/
month: '10'
oa_version: Published Version
page: '148'
project:
- _id: 0aacfa84-070f-11eb-9043-d7eb2c709234
call_identifier: H2020
grant_number: '819603'
name: Learning the shape of synaptic plasticity rules for neuronal architectures
and function through machine learning.
publication_identifier:
issn:
- 2663 - 337X
publication_status: published
publisher: Institute of Science and Technology Austria
related_material:
record:
- id: '9633'
relation: part_of_dissertation
status: public
status: public
supervisor:
- first_name: Tim P
full_name: Vogels, Tim P
id: CB6FF8D2-008F-11EA-8E08-2637E6697425
last_name: Vogels
orcid: 0000-0003-3295-6181
title: 'Synapseek: Meta-learning synaptic plasticity rules'
tmp:
image: /images/cc_by_nc_sa.png
legal_code_url: https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode
name: Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC
BY-NC-SA 4.0)
short: CC BY-NC-SA (4.0)
type: dissertation
user_id: 8b945eb4-e2f2-11eb-945a-df72226e66a9
year: '2023'
...
---
_id: '14374'
abstract:
- lang: eng
text: "Superconductivity has many important applications ranging from levitating
trains over qubits to MRI scanners. The phenomenon is successfully modeled by
Bardeen-Cooper-Schrieffer (BCS) theory. From a mathematical perspective, BCS theory
has been studied extensively for systems without boundary. However, little is
known in the presence of boundaries. With the help of numerical methods physicists
observed that the critical temperature may increase in the presence of a boundary.
The goal of this thesis is to understand the influence of boundaries on the critical
temperature in BCS theory and to give a first rigorous justification of these
observations. On the way, we also study two-body Schrödinger operators on domains
with boundaries and prove additional results for superconductors without boundary.\r\n\r\nBCS
theory is based on a non-linear functional, where the minimizer indicates whether
the system is superconducting or in the normal, non-superconducting state. By
considering the Hessian of the BCS functional at the normal state, one can analyze
whether the normal state is possibly a minimum of the BCS functional and estimate
the critical temperature. The Hessian turns out to be a linear operator resembling
a Schrödinger operator for two interacting particles, but with more complicated
kinetic energy. As a first step, we study the two-body Schrödinger operator in
the presence of boundaries.\r\nFor Neumann boundary conditions, we prove that
the addition of a boundary can create new eigenvalues, which correspond to the
two particles forming a bound state close to the boundary.\r\n\r\nSecond, we need
to understand superconductivity in the translation invariant setting. While in
three dimensions this has been extensively studied, there is no mathematical literature
for the one and two dimensional cases. In dimensions one and two, we compute the
weak coupling asymptotics of the critical temperature and the energy gap in the
translation invariant setting. We also prove that their ratio is independent of
the microscopic details of the model in the weak coupling limit; this property
is referred to as universality.\r\n\r\nIn the third part, we study the critical
temperature of superconductors in the presence of boundaries. We start by considering
the one-dimensional case of a half-line with contact interaction. Then, we generalize
the results to generic interactions and half-spaces in one, two and three dimensions.
Finally, we compare the critical temperature of a quarter space in two dimensions
to the critical temperatures of a half-space and of the full space."
alternative_title:
- ISTA Thesis
article_processing_charge: No
author:
- first_name: Barbara
full_name: Roos, Barbara
id: 5DA90512-D80F-11E9-8994-2E2EE6697425
last_name: Roos
orcid: 0000-0002-9071-5880
citation:
ama: Roos B. Boundary superconductivity in BCS theory. 2023. doi:10.15479/at:ista:14374
apa: Roos, B. (2023). Boundary superconductivity in BCS theory. Institute
of Science and Technology Austria. https://doi.org/10.15479/at:ista:14374
chicago: Roos, Barbara. “Boundary Superconductivity in BCS Theory.” Institute of
Science and Technology Austria, 2023. https://doi.org/10.15479/at:ista:14374.
ieee: B. Roos, “Boundary superconductivity in BCS theory,” Institute of Science
and Technology Austria, 2023.
ista: Roos B. 2023. Boundary superconductivity in BCS theory. Institute of Science
and Technology Austria.
mla: Roos, Barbara. Boundary Superconductivity in BCS Theory. Institute of
Science and Technology Austria, 2023, doi:10.15479/at:ista:14374.
short: B. Roos, Boundary Superconductivity in BCS Theory, Institute of Science and
Technology Austria, 2023.
date_created: 2023-09-28T14:23:04Z
date_published: 2023-09-30T00:00:00Z
date_updated: 2023-10-27T10:37:30Z
day: '30'
ddc:
- '515'
- '539'
degree_awarded: PhD
department:
- _id: GradSch
- _id: RoSe
doi: 10.15479/at:ista:14374
ec_funded: 1
file:
- access_level: open_access
checksum: ef039ffc3de2cb8dee5b14110938e9b6
content_type: application/pdf
creator: broos
date_created: 2023-10-06T11:35:56Z
date_updated: 2023-10-06T11:35:56Z
file_id: '14398'
file_name: phd-thesis-draft_pdfa_acrobat.pdf
file_size: 2365702
relation: main_file
- access_level: closed
checksum: 81dcac33daeefaf0111db52f41bb1fd0
content_type: application/x-zip-compressed
creator: broos
date_created: 2023-10-06T11:38:01Z
date_updated: 2023-10-06T11:38:01Z
file_id: '14399'
file_name: Version5.zip
file_size: 4691734
relation: source_file
file_date_updated: 2023-10-06T11:38:01Z
has_accepted_license: '1'
language:
- iso: eng
month: '09'
oa: 1
oa_version: Published Version
page: '206'
project:
- _id: 25C6DC12-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '694227'
name: Analysis of quantum many-body systems
- _id: bda63fe5-d553-11ed-ba76-a16e3d2f256b
grant_number: I06427
name: Mathematical Challenges in BCS Theory of Superconductivity
publication_identifier:
issn:
- 2663 - 337X
publication_status: published
publisher: Institute of Science and Technology Austria
related_material:
record:
- id: '13207'
relation: part_of_dissertation
status: public
- id: '10850'
relation: part_of_dissertation
status: public
status: public
supervisor:
- first_name: Robert
full_name: Seiringer, Robert
id: 4AFD0470-F248-11E8-B48F-1D18A9856A87
last_name: Seiringer
orcid: 0000-0002-6781-0521
title: Boundary superconductivity in BCS theory
tmp:
image: /images/cc_by_nc_sa.png
legal_code_url: https://creativecommons.org/licenses/by-nc-sa/4.0/legalcode
name: Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC
BY-NC-SA 4.0)
short: CC BY-NC-SA (4.0)
type: dissertation
user_id: 8b945eb4-e2f2-11eb-945a-df72226e66a9
year: '2023'
...
---
_id: '13207'
abstract:
- lang: eng
text: We consider the linear BCS equation, determining the BCS critical temperature,
in the presence of a boundary, where Dirichlet boundary conditions are imposed.
In the one-dimensional case with point interactions, we prove that the critical
temperature is strictly larger than the bulk value, at least at weak coupling.
In particular, the Cooper-pair wave function localizes near the boundary, an effect
that cannot be modeled by effective Neumann boundary conditions on the order parameter
as often imposed in Ginzburg–Landau theory. We also show that the relative shift
in critical temperature vanishes if the coupling constant either goes to zero
or to infinity.
acknowledgement: We thank Egor Babaev for encouraging us to study this problem, and
Rupert Frank for many fruitful discussions. scussions. Funding. Funding from the
European Union’s Horizon 2020 research and innovation programme under the ERC grant
agreement No. 694227 (Barbara Roos and Robert Seiringer) is gratefully acknowledged.
article_processing_charge: No
article_type: original
author:
- first_name: Christian
full_name: Hainzl, Christian
last_name: Hainzl
- first_name: Barbara
full_name: Roos, Barbara
id: 5DA90512-D80F-11E9-8994-2E2EE6697425
last_name: Roos
orcid: 0000-0002-9071-5880
- first_name: Robert
full_name: Seiringer, Robert
id: 4AFD0470-F248-11E8-B48F-1D18A9856A87
last_name: Seiringer
orcid: 0000-0002-6781-0521
citation:
ama: Hainzl C, Roos B, Seiringer R. Boundary superconductivity in the BCS model.
Journal of Spectral Theory. 2023;12(4):1507–1540. doi:10.4171/JST/439
apa: Hainzl, C., Roos, B., & Seiringer, R. (2023). Boundary superconductivity
in the BCS model. Journal of Spectral Theory. EMS Press. https://doi.org/10.4171/JST/439
chicago: Hainzl, Christian, Barbara Roos, and Robert Seiringer. “Boundary Superconductivity
in the BCS Model.” Journal of Spectral Theory. EMS Press, 2023. https://doi.org/10.4171/JST/439.
ieee: C. Hainzl, B. Roos, and R. Seiringer, “Boundary superconductivity in the BCS
model,” Journal of Spectral Theory, vol. 12, no. 4. EMS Press, pp. 1507–1540,
2023.
ista: Hainzl C, Roos B, Seiringer R. 2023. Boundary superconductivity in the BCS
model. Journal of Spectral Theory. 12(4), 1507–1540.
mla: Hainzl, Christian, et al. “Boundary Superconductivity in the BCS Model.” Journal
of Spectral Theory, vol. 12, no. 4, EMS Press, 2023, pp. 1507–1540, doi:10.4171/JST/439.
short: C. Hainzl, B. Roos, R. Seiringer, Journal of Spectral Theory 12 (2023) 1507–1540.
date_created: 2023-07-10T16:35:45Z
date_published: 2023-05-18T00:00:00Z
date_updated: 2023-10-27T10:37:29Z
day: '18'
ddc:
- '530'
department:
- _id: GradSch
- _id: RoSe
doi: 10.4171/JST/439
ec_funded: 1
external_id:
arxiv:
- '2201.08090'
isi:
- '000997933500008'
file:
- access_level: open_access
checksum: 5501da33be010b5c81440438287584d5
content_type: application/pdf
creator: alisjak
date_created: 2023-07-11T08:19:15Z
date_updated: 2023-07-11T08:19:15Z
file_id: '13208'
file_name: 2023_EMS_Hainzl.pdf
file_size: 304619
relation: main_file
success: 1
file_date_updated: 2023-07-11T08:19:15Z
has_accepted_license: '1'
intvolume: ' 12'
isi: 1
issue: '4'
language:
- iso: eng
month: '05'
oa: 1
oa_version: Published Version
page: 1507–1540
project:
- _id: 25C6DC12-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '694227'
name: Analysis of quantum many-body systems
publication: Journal of Spectral Theory
publication_identifier:
eissn:
- 1664-0403
issn:
- 1664-039X
publication_status: published
publisher: EMS Press
quality_controlled: '1'
related_material:
record:
- id: '14374'
relation: dissertation_contains
status: public
status: public
title: Boundary superconductivity in the BCS model
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: journal_article
user_id: 4359f0d1-fa6c-11eb-b949-802e58b17ae8
volume: 12
year: '2023'
...
---
_id: '14452'
abstract:
- lang: eng
text: The classical infinitesimal model is a simple and robust model for the inheritance
of quantitative traits. In this model, a quantitative trait is expressed as the
sum of a genetic and an environmental component, and the genetic component of
offspring traits within a family follows a normal distribution around the average
of the parents’ trait values, and has a variance that is independent of the parental
traits. In previous work, we showed that when trait values are determined by the
sum of a large number of additive Mendelian factors, each of small effect, one
can justify the infinitesimal model as a limit of Mendelian inheritance. In this
paper, we show that this result extends to include dominance. We define the model
in terms of classical quantities of quantitative genetics, before justifying it
as a limit of Mendelian inheritance as the number, M, of underlying loci tends
to infinity. As in the additive case, the multivariate normal distribution of
trait values across the pedigree can be expressed in terms of variance components
in an ancestral population and probabilities of identity by descent determined
by the pedigree. Now, with just first-order dominance effects, we require two-,
three-, and four-way identities. We also show that, even if we condition on parental
trait values, the “shared” and “residual” components of trait values within each
family will be asymptotically normally distributed as the number of loci tends
to infinity, with an error of order 1/M−−√. We illustrate our results with some
numerical examples.
acknowledgement: NHB was supported in part by ERC Grants 250152 and 101055327. AV
was partly supported by the chaire Modélisation Mathématique et Biodiversité of
Veolia Environment—Ecole Polytechnique—Museum National d’Histoire Naturelle—Fondation
X.
article_number: iyad133
article_processing_charge: Yes (in subscription journal)
article_type: original
author:
- first_name: Nicholas H
full_name: Barton, Nicholas H
id: 4880FE40-F248-11E8-B48F-1D18A9856A87
last_name: Barton
orcid: 0000-0002-8548-5240
- first_name: Alison M.
full_name: Etheridge, Alison M.
last_name: Etheridge
- first_name: Amandine
full_name: Véber, Amandine
last_name: Véber
citation:
ama: Barton NH, Etheridge AM, Véber A. The infinitesimal model with dominance. Genetics.
2023;225(2). doi:10.1093/genetics/iyad133
apa: Barton, N. H., Etheridge, A. M., & Véber, A. (2023). The infinitesimal
model with dominance. Genetics. Oxford Academic. https://doi.org/10.1093/genetics/iyad133
chicago: Barton, Nicholas H, Alison M. Etheridge, and Amandine Véber. “The Infinitesimal
Model with Dominance.” Genetics. Oxford Academic, 2023. https://doi.org/10.1093/genetics/iyad133.
ieee: N. H. Barton, A. M. Etheridge, and A. Véber, “The infinitesimal model with
dominance,” Genetics, vol. 225, no. 2. Oxford Academic, 2023.
ista: Barton NH, Etheridge AM, Véber A. 2023. The infinitesimal model with dominance.
Genetics. 225(2), iyad133.
mla: Barton, Nicholas H., et al. “The Infinitesimal Model with Dominance.” Genetics,
vol. 225, no. 2, iyad133, Oxford Academic, 2023, doi:10.1093/genetics/iyad133.
short: N.H. Barton, A.M. Etheridge, A. Véber, Genetics 225 (2023).
date_created: 2023-10-29T23:01:15Z
date_published: 2023-10-01T00:00:00Z
date_updated: 2023-10-30T13:04:11Z
day: '01'
ddc:
- '570'
department:
- _id: NiBa
doi: 10.1093/genetics/iyad133
ec_funded: 1
external_id:
arxiv:
- '2211.03515'
file:
- access_level: open_access
checksum: 3f65b1fbe813e2f4dbb5d2b5e891844a
content_type: application/pdf
creator: dernst
date_created: 2023-10-30T12:57:53Z
date_updated: 2023-10-30T12:57:53Z
file_id: '14469'
file_name: 2023_Genetics_Barton.pdf
file_size: 1439032
relation: main_file
success: 1
file_date_updated: 2023-10-30T12:57:53Z
has_accepted_license: '1'
intvolume: ' 225'
issue: '2'
language:
- iso: eng
month: '10'
oa: 1
oa_version: Published Version
project:
- _id: 25B07788-B435-11E9-9278-68D0E5697425
call_identifier: FP7
grant_number: '250152'
name: Limits to selection in biology and in evolutionary computation
- _id: bd6958e0-d553-11ed-ba76-86eba6a76c00
grant_number: '101055327'
name: Understanding the evolution of continuous genomes
publication: Genetics
publication_identifier:
eissn:
- 1943-2631
issn:
- 0016-6731
publication_status: published
publisher: Oxford Academic
quality_controlled: '1'
related_material:
record:
- id: '12949'
relation: research_data
status: public
scopus_import: '1'
status: public
title: The infinitesimal model with dominance
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 225
year: '2023'
...
---
_id: '12949'
abstract:
- lang: eng
text: The classical infinitesimal model is a simple and robust model for the inheritance
of quantitative traits. In this model, a quantitative trait is expressed as the
sum of a genetic and a non-genetic (environmental) component and the genetic component
of offspring traits within a family follows a normal distribution around the average
of the parents’ trait values, and has a variance that is independent of the trait
values of the parents. Although the trait distribution across the whole population
can be far from normal, the trait distributions within families are normally distributed
with a variance-covariance matrix that is determined entirely by that in the
ancestral population and the probabilities of identity determined by the pedigree.
Moreover, conditioning on some of the trait values within the pedigree has predictable
effects on the mean and variance within and between families. In previous work,
Barton et al. (2017), we showed that when trait values are determined by the sum
of a large number of Mendelian factors, each of small effect, one can justify
the infinitesimal model as limit of Mendelian inheritance. It was also shown that
under some forms of epistasis, trait values within a family are still normally
distributed.
article_processing_charge: No
author:
- first_name: Nicholas H
full_name: Barton, Nicholas H
id: 4880FE40-F248-11E8-B48F-1D18A9856A87
last_name: Barton
orcid: 0000-0002-8548-5240
citation:
ama: Barton NH. The infinitesimal model with dominance. 2023. doi:10.15479/AT:ISTA:12949
apa: Barton, N. H. (2023). The infinitesimal model with dominance. Institute of
Science and Technology Austria. https://doi.org/10.15479/AT:ISTA:12949
chicago: Barton, Nicholas H. “The Infinitesimal Model with Dominance.” Institute
of Science and Technology Austria, 2023. https://doi.org/10.15479/AT:ISTA:12949.
ieee: N. H. Barton, “The infinitesimal model with dominance.” Institute of Science
and Technology Austria, 2023.
ista: Barton NH. 2023. The infinitesimal model with dominance, Institute of Science
and Technology Austria, 10.15479/AT:ISTA:12949.
mla: Barton, Nicholas H. The Infinitesimal Model with Dominance. Institute
of Science and Technology Austria, 2023, doi:10.15479/AT:ISTA:12949.
short: N.H. Barton, (2023).
contributor:
- contributor_type: researcher
first_name: Amandine
last_name: Veber
- contributor_type: researcher
first_name: Alison
last_name: Etheridge
date_created: 2023-05-13T09:49:09Z
date_published: 2023-05-13T00:00:00Z
date_updated: 2023-10-30T13:04:11Z
day: '13'
ddc:
- '576'
department:
- _id: NiBa
doi: 10.15479/AT:ISTA:12949
file:
- access_level: open_access
checksum: b0ce7d4b1ee7e7265430ceed36fc3336
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:36:33Z
date_updated: 2023-05-13T09:36:33Z
file_id: '12950'
file_name: Neutral identities 16th Jan
file_size: 13662
relation: main_file
success: 1
- access_level: open_access
checksum: ad5035ad4f7d3b150a252c79884f6a83
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:38:17Z
date_updated: 2023-05-13T09:38:17Z
file_id: '12951'
file_name: p, zA, zD, N=30 neutral III
file_size: 181619928
relation: main_file
success: 1
- access_level: open_access
checksum: 62182a1de796256edd6f4223704312ef
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:41:59Z
date_updated: 2023-05-13T09:41:59Z
file_id: '12952'
file_name: p, zA, zD, N=30 neutral IV
file_size: 605902074
relation: main_file
success: 1
- access_level: open_access
checksum: af775dda5c4f6859cb1e5a81ec40a667
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:46:52Z
date_updated: 2023-05-13T09:46:52Z
file_id: '12953'
file_name: p, zA, zD, N=30 selected k=5
file_size: 1018238746
relation: main_file
success: 1
- access_level: open_access
checksum: af26f3394c387d3ada14b434cd68b1e5
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:42:05Z
date_updated: 2023-05-13T09:42:05Z
file_id: '12954'
file_name: Pairwise F N=30 neutral II
file_size: 3197160
relation: main_file
success: 1
- access_level: open_access
checksum: d5da7dc0e7282dd48222e26d12e34220
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:42:06Z
date_updated: 2023-05-13T09:42:06Z
file_id: '12955'
file_name: Pedigrees N=30 neutral II
file_size: 55492
relation: main_file
success: 1
- access_level: open_access
checksum: 00f386d80677590e29f6235d49cba58d
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:46:06Z
date_updated: 2023-05-13T09:46:06Z
file_id: '12956'
file_name: selected reps N=30 selected k=1,2 300 reps III
file_size: 474003467
relation: main_file
success: 1
- access_level: open_access
checksum: 658cef3eaea6136a4d24da4f074191d7
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:46:08Z
date_updated: 2023-05-13T09:46:08Z
file_id: '12957'
file_name: Algorithm for caclulating identities.nb
file_size: 121209
relation: main_file
success: 1
- access_level: open_access
checksum: db9b6dddd7a596d974e25f5e78f5c45c
content_type: application/octet-stream
creator: nbarton
date_created: 2023-05-13T09:46:08Z
date_updated: 2023-05-13T09:46:08Z
file_id: '12958'
file_name: Infinitesimal with dominance.nb
file_size: 1803898
relation: main_file
success: 1
- access_level: open_access
checksum: 91f80a9fb58cae8eef2d8bf59fe30189
content_type: text/plain
creator: nbarton
date_created: 2023-05-16T04:09:08Z
date_updated: 2023-05-16T04:09:08Z
file_id: '12967'
file_name: ReadMe.txt
file_size: 990
relation: main_file
success: 1
file_date_updated: 2023-05-16T04:09:08Z
has_accepted_license: '1'
keyword:
- Quantitative genetics
- infinitesimal model
month: '05'
oa: 1
oa_version: Published Version
project:
- _id: bd6958e0-d553-11ed-ba76-86eba6a76c00
grant_number: '101055327'
name: Understanding the evolution of continuous genomes
publisher: Institute of Science and Technology Austria
related_material:
record:
- id: '14452'
relation: used_in_publication
status: public
status: public
title: The infinitesimal model with dominance
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: research_data
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
year: '2023'
...
---
_id: '14461'
abstract:
- lang: eng
text: 'Communication-reduction techniques are a popular way to improve scalability
in data-parallel training of deep neural networks (DNNs). The recent emergence
of large language models such as GPT has created the need for new approaches to
exploit data-parallelism. Among these, fully-sharded data parallel (FSDP) training
is highly popular, yet it still encounters scalability bottlenecks. One reason
is that applying compression techniques to FSDP is challenging: as the vast majority
of the communication involves the model’s weights, direct compression alters convergence
and leads to accuracy loss. We present QSDP, a variant of FSDP which supports
both gradient and weight quantization with theoretical guarantees, is simple to
implement and has essentially no overheads. To derive QSDP we prove that a natural
modification of SGD achieves convergence even when we only maintain quantized
weights, and thus the domain over which we train consists of quantized points
and is, therefore, highly non-convex. We validate this approach by training GPT-family
models with up to 1.3 billion parameters on a multi-node cluster. Experiments
show that QSDP preserves model accuracy, while completely removing the communication
bottlenecks of FSDP, providing end-to-end speedups of up to 2.2x.'
acknowledged_ssus:
- _id: ScienComp
acknowledgement: The authors gratefully acknowledge funding from the European Research
Council (ERC) under the European Union’s Horizon 2020 research and innovation programme
(grant agreement No 805223 ScaleML), as well as experimental support from the IST
Austria IT department, in particular Stefano Elefante, Andrei Hornoiu, and Alois
Schloegl. AV acknowledges the support of the French Agence Nationale de la Recherche
(ANR), under grant ANR-21-CE48-0016 (project COMCOPT), the support of Fondation
Hadamard with a PRMO grant, and the support of CNRS with a CoopIntEER IEA grant
(project ALFRED).
alternative_title:
- PMLR
article_processing_charge: No
author:
- first_name: Ilia
full_name: Markov, Ilia
id: D0CF4148-C985-11E9-8066-0BDEE5697425
last_name: Markov
- first_name: Adrian
full_name: Vladu, Adrian
last_name: Vladu
- first_name: Qi
full_name: Guo, Qi
last_name: Guo
- first_name: Dan-Adrian
full_name: Alistarh, Dan-Adrian
id: 4A899BFC-F248-11E8-B48F-1D18A9856A87
last_name: Alistarh
orcid: 0000-0003-3650-940X
citation:
ama: 'Markov I, Vladu A, Guo Q, Alistarh D-A. Quantized distributed training of
large models with convergence guarantees. In: Proceedings of the 40th International
Conference on Machine Learning. Vol 202. ML Research Press; 2023:24020-24044.'
apa: 'Markov, I., Vladu, A., Guo, Q., & Alistarh, D.-A. (2023). Quantized distributed
training of large models with convergence guarantees. In Proceedings of the
40th International Conference on Machine Learning (Vol. 202, pp. 24020–24044).
Honolulu, Hawaii, HI, United States: ML Research Press.'
chicago: Markov, Ilia, Adrian Vladu, Qi Guo, and Dan-Adrian Alistarh. “Quantized
Distributed Training of Large Models with Convergence Guarantees.” In Proceedings
of the 40th International Conference on Machine Learning, 202:24020–44. ML
Research Press, 2023.
ieee: I. Markov, A. Vladu, Q. Guo, and D.-A. Alistarh, “Quantized distributed training
of large models with convergence guarantees,” in Proceedings of the 40th International
Conference on Machine Learning, Honolulu, Hawaii, HI, United States, 2023,
vol. 202, pp. 24020–24044.
ista: 'Markov I, Vladu A, Guo Q, Alistarh D-A. 2023. Quantized distributed training
of large models with convergence guarantees. Proceedings of the 40th International
Conference on Machine Learning. ICML: International Conference on Machine Learning,
PMLR, vol. 202, 24020–24044.'
mla: Markov, Ilia, et al. “Quantized Distributed Training of Large Models with Convergence
Guarantees.” Proceedings of the 40th International Conference on Machine Learning,
vol. 202, ML Research Press, 2023, pp. 24020–44.
short: I. Markov, A. Vladu, Q. Guo, D.-A. Alistarh, in:, Proceedings of the 40th
International Conference on Machine Learning, ML Research Press, 2023, pp. 24020–24044.
conference:
end_date: 2023-07-29
location: Honolulu, Hawaii, HI, United States
name: 'ICML: International Conference on Machine Learning'
start_date: 2023-07-23
date_created: 2023-10-29T23:01:17Z
date_published: 2023-07-30T00:00:00Z
date_updated: 2023-10-31T09:40:45Z
day: '30'
department:
- _id: DaAl
ec_funded: 1
external_id:
arxiv:
- '2302.02390'
intvolume: ' 202'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://doi.org/10.48550/arXiv.2302.02390
month: '07'
oa: 1
oa_version: Preprint
page: 24020-24044
project:
- _id: 268A44D6-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '805223'
name: Elastic Coordination for Scalable Machine Learning
publication: Proceedings of the 40th International Conference on Machine Learning
publication_identifier:
eissn:
- 2640-3498
publication_status: published
publisher: ML Research Press
quality_controlled: '1'
scopus_import: '1'
status: public
title: Quantized distributed training of large models with convergence guarantees
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 202
year: '2023'
...
---
_id: '14462'
abstract:
- lang: eng
text: "We study fine-grained error bounds for differentially private algorithms
for counting under continual observation. Our main insight is that the matrix
mechanism when using lower-triangular matrices can be used in the continual observation
model. More specifically, we give an explicit factorization for the counting matrix
Mcount and upper bound the error explicitly. We also give a fine-grained analysis,
specifying the exact constant in the upper bound. Our analysis is based on upper
and lower bounds of the completely bounded norm (cb-norm) of Mcount\r\n. Along
the way, we improve the best-known bound of 28 years by Mathias (SIAM Journal
on Matrix Analysis and Applications, 1993) on the cb-norm of Mcount for a large
range of the dimension of Mcount. Furthermore, we are the first to give concrete
error bounds for various problems under continual observation such as binary counting,
maintaining a histogram, releasing an approximately cut-preserving synthetic graph,
many graph-based statistics, and substring and episode counting. Finally, we note
that our result can be used to get a fine-grained error bound for non-interactive
local learning and the first lower bounds on the additive error for (ϵ,δ)-differentially-private
counting under continual observation. Subsequent to this work, Henzinger et al.
(SODA, 2023) showed that our factorization also achieves fine-grained mean-squared
error."
acknowledgement: "This project has received funding from the European Research Council
(ERC) under the European Union’s Horizon 2020 research and innovation programme
(Grant agreement No.\r\n101019564 “The Design of Modern Fully Dynamic Data Structures
(MoDynStruct)” and from the Austrian Science Fund (FWF) project Z 422-N, and project
“Fast Algorithms for a Reactive Network Layer (ReactNet)”, P 33775-N, with additional
funding from the netidee SCIENCE Stiftung, 2020–2024. 2020–2024. JU’s research was
funded by Decanal Research Grant. A part of this work was done when JU was visiting
Indian Statistical Institute, Delhi. The authors would like to thank Rajat Bhatia,
Aleksandar Nikolov, Shanta Laisharam, Vern Paulsen, Ryan Rogers, Abhradeep Thakurta,
and Sarvagya Upadhyay for useful discussions."
alternative_title:
- PMLR
article_processing_charge: No
author:
- first_name: Hendrik
full_name: Fichtenberger, Hendrik
last_name: Fichtenberger
- first_name: Monika H
full_name: Henzinger, Monika H
id: 540c9bbd-f2de-11ec-812d-d04a5be85630
last_name: Henzinger
orcid: 0000-0002-5008-6530
- first_name: Jalaj
full_name: Upadhyay, Jalaj
last_name: Upadhyay
citation:
ama: 'Fichtenberger H, Henzinger MH, Upadhyay J. Constant matters: Fine-grained
error bound on differentially private continual observation. In: Proceedings
of the 40th International Conference on Machine Learning. Vol 202. ML Research
Press; 2023:10072-10092.'
apa: 'Fichtenberger, H., Henzinger, M. H., & Upadhyay, J. (2023). Constant matters:
Fine-grained error bound on differentially private continual observation. In Proceedings
of the 40th International Conference on Machine Learning (Vol. 202, pp. 10072–10092).
Honolulu, Hawaii, HI, United States: ML Research Press.'
chicago: 'Fichtenberger, Hendrik, Monika H Henzinger, and Jalaj Upadhyay. “Constant
Matters: Fine-Grained Error Bound on Differentially Private Continual Observation.”
In Proceedings of the 40th International Conference on Machine Learning,
202:10072–92. ML Research Press, 2023.'
ieee: 'H. Fichtenberger, M. H. Henzinger, and J. Upadhyay, “Constant matters: Fine-grained
error bound on differentially private continual observation,” in Proceedings
of the 40th International Conference on Machine Learning, Honolulu, Hawaii,
HI, United States, 2023, vol. 202, pp. 10072–10092.'
ista: 'Fichtenberger H, Henzinger MH, Upadhyay J. 2023. Constant matters: Fine-grained
error bound on differentially private continual observation. Proceedings of the
40th International Conference on Machine Learning. ICML: International Conference
on Machine Learning, PMLR, vol. 202, 10072–10092.'
mla: 'Fichtenberger, Hendrik, et al. “Constant Matters: Fine-Grained Error Bound
on Differentially Private Continual Observation.” Proceedings of the 40th International
Conference on Machine Learning, vol. 202, ML Research Press, 2023, pp. 10072–92.'
short: H. Fichtenberger, M.H. Henzinger, J. Upadhyay, in:, Proceedings of the 40th
International Conference on Machine Learning, ML Research Press, 2023, pp. 10072–10092.
conference:
end_date: 2023-07-29
location: Honolulu, Hawaii, HI, United States
name: 'ICML: International Conference on Machine Learning'
start_date: 2023-07-23
date_created: 2023-10-29T23:01:17Z
date_published: 2023-07-30T00:00:00Z
date_updated: 2023-10-31T09:54:05Z
day: '30'
department:
- _id: MoHe
ec_funded: 1
intvolume: ' 202'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://proceedings.mlr.press/v202/fichtenberger23a/fichtenberger23a.pdf
month: '07'
oa: 1
oa_version: Published Version
page: 10072-10092
project:
- _id: bd9ca328-d553-11ed-ba76-dc4f890cfe62
call_identifier: H2020
grant_number: '101019564'
name: The design and evaluation of modern fully dynamic data structures
- _id: 34def286-11ca-11ed-8bc3-da5948e1613c
grant_number: Z00422
name: Wittgenstein Award - Monika Henzinger
- _id: bd9e3a2e-d553-11ed-ba76-8aa684ce17fe
grant_number: 'P33775 '
name: Fast Algorithms for a Reactive Network Layer
publication: Proceedings of the 40th International Conference on Machine Learning
publication_identifier:
eissn:
- 2640-3498
publication_status: published
publisher: ML Research Press
quality_controlled: '1'
scopus_import: '1'
status: public
title: 'Constant matters: Fine-grained error bound on differentially private continual
observation'
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 202
year: '2023'
...
---
_id: '14459'
abstract:
- lang: eng
text: Autoencoders are a popular model in many branches of machine learning and
lossy data compression. However, their fundamental limits, the performance of
gradient methods and the features learnt during optimization remain poorly understood,
even in the two-layer setting. In fact, earlier work has considered either linear
autoencoders or specific training regimes (leading to vanishing or diverging compression
rates). Our paper addresses this gap by focusing on non-linear two-layer autoencoders
trained in the challenging proportional regime in which the input dimension scales
linearly with the size of the representation. Our results characterize the minimizers
of the population risk, and show that such minimizers are achieved by gradient
methods; their structure is also unveiled, thus leading to a concise description
of the features obtained via training. For the special case of a sign activation
function, our analysis establishes the fundamental limits for the lossy compression
of Gaussian sources via (shallow) autoencoders. Finally, while the results are
proved for Gaussian data, numerical simulations on standard datasets display the
universality of the theoretical predictions.
acknowledgement: Aleksandr Shevchenko, Kevin Kogler and Marco Mondelli are supported
by the 2019 Lopez-Loreta Prize. Hamed Hassani acknowledges the support by the NSF
CIF award (1910056) and the NSF Institute for CORE Emerging Methods in Data Science
(EnCORE).
alternative_title:
- PMLR
article_processing_charge: No
author:
- first_name: Aleksandr
full_name: Shevchenko, Aleksandr
id: F2B06EC2-C99E-11E9-89F0-752EE6697425
last_name: Shevchenko
- first_name: Kevin
full_name: Kögler, Kevin
id: 94ec913c-dc85-11ea-9058-e5051ab2428b
last_name: Kögler
- first_name: Hamed
full_name: Hassani, Hamed
last_name: Hassani
- first_name: Marco
full_name: Mondelli, Marco
id: 27EB676C-8706-11E9-9510-7717E6697425
last_name: Mondelli
orcid: 0000-0002-3242-7020
citation:
ama: 'Shevchenko A, Kögler K, Hassani H, Mondelli M. Fundamental limits of two-layer
autoencoders, and achieving them with gradient methods. In: Proceedings of
the 40th International Conference on Machine Learning. Vol 202. ML Research
Press; 2023:31151-31209.'
apa: 'Shevchenko, A., Kögler, K., Hassani, H., & Mondelli, M. (2023). Fundamental
limits of two-layer autoencoders, and achieving them with gradient methods. In
Proceedings of the 40th International Conference on Machine Learning (Vol.
202, pp. 31151–31209). Honolulu, Hawaii, HI, United States: ML Research Press.'
chicago: Shevchenko, Aleksandr, Kevin Kögler, Hamed Hassani, and Marco Mondelli.
“Fundamental Limits of Two-Layer Autoencoders, and Achieving Them with Gradient
Methods.” In Proceedings of the 40th International Conference on Machine Learning,
202:31151–209. ML Research Press, 2023.
ieee: A. Shevchenko, K. Kögler, H. Hassani, and M. Mondelli, “Fundamental limits
of two-layer autoencoders, and achieving them with gradient methods,” in Proceedings
of the 40th International Conference on Machine Learning, Honolulu, Hawaii,
HI, United States, 2023, vol. 202, pp. 31151–31209.
ista: 'Shevchenko A, Kögler K, Hassani H, Mondelli M. 2023. Fundamental limits of
two-layer autoencoders, and achieving them with gradient methods. Proceedings
of the 40th International Conference on Machine Learning. ICML: International
Conference on Machine Learning, PMLR, vol. 202, 31151–31209.'
mla: Shevchenko, Aleksandr, et al. “Fundamental Limits of Two-Layer Autoencoders,
and Achieving Them with Gradient Methods.” Proceedings of the 40th International
Conference on Machine Learning, vol. 202, ML Research Press, 2023, pp. 31151–209.
short: A. Shevchenko, K. Kögler, H. Hassani, M. Mondelli, in:, Proceedings of the
40th International Conference on Machine Learning, ML Research Press, 2023, pp.
31151–31209.
conference:
end_date: 2023-07-29
location: Honolulu, Hawaii, HI, United States
name: 'ICML: International Conference on Machine Learning'
start_date: 2023-07-23
date_created: 2023-10-29T23:01:17Z
date_published: 2023-07-30T00:00:00Z
date_updated: 2023-10-31T08:52:28Z
day: '30'
department:
- _id: MaMo
- _id: DaAl
external_id:
arxiv:
- '2212.13468'
intvolume: ' 202'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://doi.org/10.48550/arXiv.2212.13468
month: '07'
oa: 1
oa_version: Preprint
page: 31151-31209
project:
- _id: 059876FA-7A3F-11EA-A408-12923DDC885E
name: Prix Lopez-Loretta 2019 - Marco Mondelli
publication: Proceedings of the 40th International Conference on Machine Learning
publication_identifier:
eissn:
- 2640-3498
publication_status: published
publisher: ML Research Press
quality_controlled: '1'
scopus_import: '1'
status: public
title: Fundamental limits of two-layer autoencoders, and achieving them with gradient
methods
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 202
year: '2023'
...
---
_id: '14460'
abstract:
- lang: eng
text: We provide an efficient implementation of the backpropagation algorithm, specialized
to the case where the weights of the neural network being trained are sparse.
Our algorithm is general, as it applies to arbitrary (unstructured) sparsity and
common layer types (e.g., convolutional or linear). We provide a fast vectorized
implementation on commodity CPUs, and show that it can yield speedups in end-to-end
runtime experiments, both in transfer learning using already-sparsified networks,
and in training sparse networks from scratch. Thus, our results provide the first
support for sparse training on commodity hardware.
acknowledgement: 'We would like to thank Elias Frantar for his valuable assistance
and support at the outset of this project, and the anonymous ICML and SNN reviewers
for very constructive feedback. EI was supported in part by the FWF DK VGSCO, grant
agreement number W1260-N35. DA acknowledges generous ERC support, via Starting Grant
805223 ScaleML. '
alternative_title:
- PMLR
article_processing_charge: No
author:
- first_name: Mahdi
full_name: Nikdan, Mahdi
id: 66374281-f394-11eb-9cf6-869147deecc0
last_name: Nikdan
- first_name: Tommaso
full_name: Pegolotti, Tommaso
last_name: Pegolotti
- first_name: Eugenia B
full_name: Iofinova, Eugenia B
id: f9a17499-f6e0-11ea-865d-fdf9a3f77117
last_name: Iofinova
orcid: 0000-0002-7778-3221
- first_name: Eldar
full_name: Kurtic, Eldar
id: 47beb3a5-07b5-11eb-9b87-b108ec578218
last_name: Kurtic
- first_name: Dan-Adrian
full_name: Alistarh, Dan-Adrian
id: 4A899BFC-F248-11E8-B48F-1D18A9856A87
last_name: Alistarh
orcid: 0000-0003-3650-940X
citation:
ama: 'Nikdan M, Pegolotti T, Iofinova EB, Kurtic E, Alistarh D-A. SparseProp: Efficient
sparse backpropagation for faster training of neural networks at the edge. In:
Proceedings of the 40th International Conference on Machine Learning. Vol
202. ML Research Press; 2023:26215-26227.'
apa: 'Nikdan, M., Pegolotti, T., Iofinova, E. B., Kurtic, E., & Alistarh, D.-A.
(2023). SparseProp: Efficient sparse backpropagation for faster training of neural
networks at the edge. In Proceedings of the 40th International Conference on
Machine Learning (Vol. 202, pp. 26215–26227). Honolulu, Hawaii, HI, United
States: ML Research Press.'
chicago: 'Nikdan, Mahdi, Tommaso Pegolotti, Eugenia B Iofinova, Eldar Kurtic, and
Dan-Adrian Alistarh. “SparseProp: Efficient Sparse Backpropagation for Faster
Training of Neural Networks at the Edge.” In Proceedings of the 40th International
Conference on Machine Learning, 202:26215–27. ML Research Press, 2023.'
ieee: 'M. Nikdan, T. Pegolotti, E. B. Iofinova, E. Kurtic, and D.-A. Alistarh, “SparseProp:
Efficient sparse backpropagation for faster training of neural networks at the
edge,” in Proceedings of the 40th International Conference on Machine Learning,
Honolulu, Hawaii, HI, United States, 2023, vol. 202, pp. 26215–26227.'
ista: 'Nikdan M, Pegolotti T, Iofinova EB, Kurtic E, Alistarh D-A. 2023. SparseProp:
Efficient sparse backpropagation for faster training of neural networks at the
edge. Proceedings of the 40th International Conference on Machine Learning. ICML:
International Conference on Machine Learning, PMLR, vol. 202, 26215–26227.'
mla: 'Nikdan, Mahdi, et al. “SparseProp: Efficient Sparse Backpropagation for Faster
Training of Neural Networks at the Edge.” Proceedings of the 40th International
Conference on Machine Learning, vol. 202, ML Research Press, 2023, pp. 26215–27.'
short: M. Nikdan, T. Pegolotti, E.B. Iofinova, E. Kurtic, D.-A. Alistarh, in:, Proceedings
of the 40th International Conference on Machine Learning, ML Research Press, 2023,
pp. 26215–26227.
conference:
end_date: 2023-07-29
location: Honolulu, Hawaii, HI, United States
name: 'ICML: International Conference on Machine Learning'
start_date: 2023-07-23
date_created: 2023-10-29T23:01:17Z
date_published: 2023-07-30T00:00:00Z
date_updated: 2023-10-31T09:33:51Z
day: '30'
department:
- _id: DaAl
ec_funded: 1
external_id:
arxiv:
- '2302.04852'
intvolume: ' 202'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://doi.org/10.48550/arXiv.2302.04852
month: '07'
oa: 1
oa_version: Preprint
page: 26215-26227
project:
- _id: 268A44D6-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '805223'
name: Elastic Coordination for Scalable Machine Learning
publication: Proceedings of the 40th International Conference on Machine Learning
publication_identifier:
eissn:
- 2640-3498
publication_status: published
publisher: ML Research Press
quality_controlled: '1'
scopus_import: '1'
status: public
title: 'SparseProp: Efficient sparse backpropagation for faster training of neural
networks at the edge'
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 202
year: '2023'
...
---
_id: '14457'
abstract:
- lang: eng
text: "Threshold secret sharing allows a dealer to split a secret s into n shares,
such that any t shares allow for reconstructing s, but no t-1 shares reveal any
information about s. Leakage-resilient secret sharing requires that the secret
remains hidden, even when an adversary additionally obtains a limited amount of
leakage from every share. Benhamouda et al. (CRYPTO’18) proved that Shamir’s secret
sharing scheme is one bit leakage-resilient for reconstruction threshold t≥0.85n
and conjectured that the same holds for t = c.n for any constant 0≤c≤1. Nielsen
and Simkin (EUROCRYPT’20) showed that this is the best one can hope for by proving
that Shamir’s scheme is not secure against one-bit leakage when t0c.n/log(n).\r\nIn
this work, we strengthen the lower bound of Nielsen and Simkin. We consider noisy
leakage-resilience, where a random subset of leakages is replaced by uniformly
random noise. We prove a lower bound for Shamir’s secret sharing, similar to that
of Nielsen and Simkin, which holds even when a constant fraction of leakages is
replaced by random noise. To this end, we first prove a lower bound on the share
size of any noisy-leakage-resilient sharing scheme. We then use this lower bound
to show that there exist universal constants c1, c2, such that for sufficiently
large n it holds that Shamir’s secret sharing scheme is not noisy-leakage-resilient
for t≤c1.n/log(n), even when a c2 fraction of leakages are replaced by random
noise.\r\n\r\n\r\n\r\n"
alternative_title:
- LNCS
article_processing_charge: No
author:
- first_name: Charlotte
full_name: Hoffmann, Charlotte
id: 0f78d746-dc7d-11ea-9b2f-83f92091afe7
last_name: Hoffmann
orcid: 0000-0003-2027-5549
- first_name: Mark
full_name: Simkin, Mark
last_name: Simkin
citation:
ama: 'Hoffmann C, Simkin M. Stronger lower bounds for leakage-resilient secret sharing.
In: 8th International Conference on Cryptology and Information Security in
Latin America. Vol 14168. Springer Nature; 2023:215-228. doi:10.1007/978-3-031-44469-2_11'
apa: 'Hoffmann, C., & Simkin, M. (2023). Stronger lower bounds for leakage-resilient
secret sharing. In 8th International Conference on Cryptology and Information
Security in Latin America (Vol. 14168, pp. 215–228). Quito, Ecuador: Springer
Nature. https://doi.org/10.1007/978-3-031-44469-2_11'
chicago: Hoffmann, Charlotte, and Mark Simkin. “Stronger Lower Bounds for Leakage-Resilient
Secret Sharing.” In 8th International Conference on Cryptology and Information
Security in Latin America, 14168:215–28. Springer Nature, 2023. https://doi.org/10.1007/978-3-031-44469-2_11.
ieee: C. Hoffmann and M. Simkin, “Stronger lower bounds for leakage-resilient secret
sharing,” in 8th International Conference on Cryptology and Information Security
in Latin America, Quito, Ecuador, 2023, vol. 14168, pp. 215–228.
ista: 'Hoffmann C, Simkin M. 2023. Stronger lower bounds for leakage-resilient secret
sharing. 8th International Conference on Cryptology and Information Security in
Latin America. LATINCRYPT: Conference on Cryptology and Information Security in
Latin America, LNCS, vol. 14168, 215–228.'
mla: Hoffmann, Charlotte, and Mark Simkin. “Stronger Lower Bounds for Leakage-Resilient
Secret Sharing.” 8th International Conference on Cryptology and Information
Security in Latin America, vol. 14168, Springer Nature, 2023, pp. 215–28,
doi:10.1007/978-3-031-44469-2_11.
short: C. Hoffmann, M. Simkin, in:, 8th International Conference on Cryptology and
Information Security in Latin America, Springer Nature, 2023, pp. 215–228.
conference:
end_date: 2023-10-06
location: Quito, Ecuador
name: 'LATINCRYPT: Conference on Cryptology and Information Security in Latin America'
start_date: 2023-10-03
date_created: 2023-10-29T23:01:16Z
date_published: 2023-10-01T00:00:00Z
date_updated: 2023-10-31T11:43:12Z
day: '01'
department:
- _id: KrPi
doi: 10.1007/978-3-031-44469-2_11
intvolume: ' 14168'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://eprint.iacr.org/2023/1017
month: '10'
oa: 1
oa_version: Preprint
page: 215-228
publication: 8th International Conference on Cryptology and Information Security in
Latin America
publication_identifier:
eissn:
- 1611-3349
isbn:
- '9783031444685'
issn:
- 0302-9743
publication_status: published
publisher: Springer Nature
quality_controlled: '1'
scopus_import: '1'
status: public
title: Stronger lower bounds for leakage-resilient secret sharing
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 14168
year: '2023'
...
---
_id: '14458'
abstract:
- lang: eng
text: 'We show for the first time that large-scale generative pretrained transformer
(GPT) family models can be pruned to at least 50% sparsity in one-shot, without
any retraining, at minimal loss of accuracy. This is achieved via a new pruning
method called SparseGPT, specifically designed to work efficiently and accurately
on massive GPT-family models. We can execute SparseGPT on the largest available
open-source models, OPT-175B and BLOOM-176B, in under 4.5 hours, and can reach
60% unstructured sparsity with negligible increase in perplexity: remarkably,
more than 100 billion weights from these models can be ignored at inference time.
SparseGPT generalizes to semi-structured (2:4 and 4:8) patterns, and is compatible
with weight quantization approaches. The code is available at: https://github.com/IST-DASLab/sparsegpt.'
acknowledged_ssus:
- _id: ScienComp
acknowledgement: The authors gratefully acknowledge funding from the European Research
Council (ERC) under the European Union’s Horizon 2020 programme (grant agreement
No. 805223 ScaleML), as well as experimental support from Eldar Kurtic, and from
the IST Austria IT department, in particular Stefano Elefante, Andrei Hornoiu, and
Alois Schloegl.
alternative_title:
- PMLR
article_processing_charge: No
author:
- first_name: Elias
full_name: Frantar, Elias
id: 09a8f98d-ec99-11ea-ae11-c063a7b7fe5f
last_name: Frantar
- first_name: Dan-Adrian
full_name: Alistarh, Dan-Adrian
id: 4A899BFC-F248-11E8-B48F-1D18A9856A87
last_name: Alistarh
orcid: 0000-0003-3650-940X
citation:
ama: 'Frantar E, Alistarh D-A. SparseGPT: Massive language models can be accurately
pruned in one-shot. In: Proceedings of the 40th International Conference on
Machine Learning. Vol 202. ML Research Press; 2023:10323-10337.'
apa: 'Frantar, E., & Alistarh, D.-A. (2023). SparseGPT: Massive language models
can be accurately pruned in one-shot. In Proceedings of the 40th International
Conference on Machine Learning (Vol. 202, pp. 10323–10337). Honolulu, Hawaii,
HI, United States: ML Research Press.'
chicago: 'Frantar, Elias, and Dan-Adrian Alistarh. “SparseGPT: Massive Language
Models Can Be Accurately Pruned in One-Shot.” In Proceedings of the 40th International
Conference on Machine Learning, 202:10323–37. ML Research Press, 2023.'
ieee: 'E. Frantar and D.-A. Alistarh, “SparseGPT: Massive language models can be
accurately pruned in one-shot,” in Proceedings of the 40th International Conference
on Machine Learning, Honolulu, Hawaii, HI, United States, 2023, vol. 202,
pp. 10323–10337.'
ista: 'Frantar E, Alistarh D-A. 2023. SparseGPT: Massive language models can be
accurately pruned in one-shot. Proceedings of the 40th International Conference
on Machine Learning. ICML: International Conference on Machine Learning, PMLR,
vol. 202, 10323–10337.'
mla: 'Frantar, Elias, and Dan-Adrian Alistarh. “SparseGPT: Massive Language Models
Can Be Accurately Pruned in One-Shot.” Proceedings of the 40th International
Conference on Machine Learning, vol. 202, ML Research Press, 2023, pp. 10323–37.'
short: E. Frantar, D.-A. Alistarh, in:, Proceedings of the 40th International Conference
on Machine Learning, ML Research Press, 2023, pp. 10323–10337.
conference:
end_date: 2023-07-29
location: Honolulu, Hawaii, HI, United States
name: 'ICML: International Conference on Machine Learning'
start_date: 2023-07-23
date_created: 2023-10-29T23:01:16Z
date_published: 2023-07-30T00:00:00Z
date_updated: 2023-10-31T09:59:42Z
day: '30'
department:
- _id: DaAl
ec_funded: 1
external_id:
arxiv:
- '2301.00774'
intvolume: ' 202'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://doi.org/10.48550/arXiv.2301.00774
month: '07'
oa: 1
oa_version: Preprint
page: 10323-10337
project:
- _id: 268A44D6-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '805223'
name: Elastic Coordination for Scalable Machine Learning
publication: Proceedings of the 40th International Conference on Machine Learning
publication_identifier:
eissn:
- 2640-3498
publication_status: published
publisher: ML Research Press
quality_controlled: '1'
scopus_import: '1'
status: public
title: 'SparseGPT: Massive language models can be accurately pruned in one-shot'
type: conference
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 202
year: '2023'
...
---
_id: '14451'
abstract:
- lang: eng
text: 'We investigate the potential of Multi-Objective, Deep Reinforcement Learning
for stock and cryptocurrency single-asset trading: in particular, we consider
a Multi-Objective algorithm which generalizes the reward functions and discount
factor (i.e., these components are not specified a priori, but incorporated in
the learning process). Firstly, using several important assets (BTCUSD, ETHUSDT,
XRPUSDT, AAPL, SPY, NIFTY50), we verify the reward generalization property of
the proposed Multi-Objective algorithm, and provide preliminary statistical evidence
showing increased predictive stability over the corresponding Single-Objective
strategy. Secondly, we show that the Multi-Objective algorithm has a clear edge
over the corresponding Single-Objective strategy when the reward mechanism is
sparse (i.e., when non-null feedback is infrequent over time). Finally, we discuss
the generalization properties with respect to the discount factor. The entirety
of our code is provided in open-source format.'
acknowledgement: Open access funding provided by Università degli Studi di Trieste
within the CRUI-CARE Agreement. Funding was provided by Austrian Science Fund (Grant
No. F65), Horizon 2020 (Grant No. 754411) and Österreichische Forschungsförderungsgesellschaft.
article_processing_charge: Yes (via OA deal)
article_type: original
author:
- first_name: Federico
full_name: Cornalba, Federico
id: 2CEB641C-A400-11E9-A717-D712E6697425
last_name: Cornalba
orcid: 0000-0002-6269-5149
- first_name: Constantin
full_name: Disselkamp, Constantin
last_name: Disselkamp
- first_name: Davide
full_name: Scassola, Davide
last_name: Scassola
- first_name: Christopher
full_name: Helf, Christopher
last_name: Helf
citation:
ama: 'Cornalba F, Disselkamp C, Scassola D, Helf C. Multi-objective reward generalization:
improving performance of Deep Reinforcement Learning for applications in single-asset
trading. Neural Computing and Applications. 2023. doi:10.1007/s00521-023-09033-7'
apa: 'Cornalba, F., Disselkamp, C., Scassola, D., & Helf, C. (2023). Multi-objective
reward generalization: improving performance of Deep Reinforcement Learning for
applications in single-asset trading. Neural Computing and Applications.
Springer Nature. https://doi.org/10.1007/s00521-023-09033-7'
chicago: 'Cornalba, Federico, Constantin Disselkamp, Davide Scassola, and Christopher
Helf. “Multi-Objective Reward Generalization: Improving Performance of Deep Reinforcement
Learning for Applications in Single-Asset Trading.” Neural Computing and Applications.
Springer Nature, 2023. https://doi.org/10.1007/s00521-023-09033-7.'
ieee: 'F. Cornalba, C. Disselkamp, D. Scassola, and C. Helf, “Multi-objective reward
generalization: improving performance of Deep Reinforcement Learning for applications
in single-asset trading,” Neural Computing and Applications. Springer Nature,
2023.'
ista: 'Cornalba F, Disselkamp C, Scassola D, Helf C. 2023. Multi-objective reward
generalization: improving performance of Deep Reinforcement Learning for applications
in single-asset trading. Neural Computing and Applications.'
mla: 'Cornalba, Federico, et al. “Multi-Objective Reward Generalization: Improving
Performance of Deep Reinforcement Learning for Applications in Single-Asset Trading.”
Neural Computing and Applications, Springer Nature, 2023, doi:10.1007/s00521-023-09033-7.'
short: F. Cornalba, C. Disselkamp, D. Scassola, C. Helf, Neural Computing and Applications
(2023).
date_created: 2023-10-22T22:01:16Z
date_published: 2023-10-05T00:00:00Z
date_updated: 2023-10-31T10:58:28Z
day: '05'
department:
- _id: JuFi
doi: 10.1007/s00521-023-09033-7
ec_funded: 1
external_id:
arxiv:
- '2203.04579'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://doi.org/10.1007/s00521-023-09033-7
month: '10'
oa: 1
oa_version: Published Version
project:
- _id: fc31cba2-9c52-11eb-aca3-ff467d239cd2
grant_number: F6504
name: Taming Complexity in Partial Differential Systems
- _id: 260C2330-B435-11E9-9278-68D0E5697425
call_identifier: H2020
grant_number: '754411'
name: ISTplus - Postdoctoral Fellowships
publication: Neural Computing and Applications
publication_identifier:
eissn:
- 1433-3058
issn:
- 0941-0643
publication_status: epub_ahead
publisher: Springer Nature
quality_controlled: '1'
scopus_import: '1'
status: public
title: 'Multi-objective reward generalization: improving performance of Deep Reinforcement
Learning for applications in single-asset trading'
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
year: '2023'
...