Learning probabilistic neural representations with randomly connected circuits

O. Maoz, G. Tkačik, M.S. Esteki, R. Kiani, E. Schneidman, Proceedings of the National Academy of Sciences of the United States of America 117 (2020) 25066–25073.

Download
OA 2020_PNAS_Maoz.pdf 1.76 MB

Journal Article | Published | English

Scopus indexed
Author
Maoz, Ori; Tkacik, GasperIST Austria ; Esteki, Mohamad Saleh; Kiani, Roozbeh; Schneidman, Elad
Department
Abstract
The brain represents and reasons probabilistically about complex stimuli and motor actions using a noisy, spike-based neural code. A key building block for such neural computations, as well as the basis for supervised and unsupervised learning, is the ability to estimate the surprise or likelihood of incoming high-dimensional neural activity patterns. Despite progress in statistical modeling of neural responses and deep learning, current approaches either do not scale to large neural populations or cannot be implemented using biologically realistic mechanisms. Inspired by the sparse and random connectivity of real neuronal circuits, we present a model for neural codes that accurately estimates the likelihood of individual spiking patterns and has a straightforward, scalable, efficient, learnable, and realistic neural implementation. This model’s performance on simultaneously recorded spiking activity of >100 neurons in the monkey visual and prefrontal cortices is comparable with or better than that of state-of-the-art models. Importantly, the model can be learned using a small number of samples and using a local learning rule that utilizes noise intrinsic to neural circuits. Slower, structural changes in random connectivity, consistent with rewiring and pruning processes, further improve the efficiency and sparseness of the resulting neural representations. Our results merge insights from neuroanatomy, machine learning, and theoretical neuroscience to suggest random sparse connectivity as a key design principle for neuronal computation.
Publishing Year
Date Published
2020-10-06
Journal Title
Proceedings of the National Academy of Sciences of the United States of America
Acknowledgement
We thank Udi Karpas, Roy Harpaz, Tal Tamir, Adam Haber, and Amir Bar for discussions and suggestions; and especially Oren Forkosh and Walter Senn for invaluable discussions of the learning rule. This work was supported by European Research Council Grant 311238 (to E.S.) and Israel Science Foundation Grant 1629/12 (to E.S.); as well as research support from Martin Kushner Schnur and Mr. and Mrs. Lawrence Feis (E.S.); National Institute of Mental Health Grant R01MH109180 (to R.K.); a Pew Scholarship in Biomedical Sciences (to R.K.); Simons Collaboration on the Global Brain Grant 542997 (to R.K. and E.S.); and a CRCNS (Collaborative Research in Computational Neuroscience) grant (to R.K. and E.S.).
Volume
117
Issue
40
Page
25066-25073
ISSN
eISSN
IST-REx-ID

Cite this

Maoz O, Tkačik G, Esteki MS, Kiani R, Schneidman E. Learning probabilistic neural representations with randomly connected circuits. Proceedings of the National Academy of Sciences of the United States of America. 2020;117(40):25066-25073. doi:10.1073/pnas.1912804117
Maoz, O., Tkačik, G., Esteki, M. S., Kiani, R., & Schneidman, E. (2020). Learning probabilistic neural representations with randomly connected circuits. Proceedings of the National Academy of Sciences of the United States of America, 117(40), 25066–25073. https://doi.org/10.1073/pnas.1912804117
Maoz, Ori, Gašper Tkačik, Mohamad Saleh Esteki, Roozbeh Kiani, and Elad Schneidman. “Learning Probabilistic Neural Representations with Randomly Connected Circuits.” Proceedings of the National Academy of Sciences of the United States of America 117, no. 40 (2020): 25066–73. https://doi.org/10.1073/pnas.1912804117.
O. Maoz, G. Tkačik, M. S. Esteki, R. Kiani, and E. Schneidman, “Learning probabilistic neural representations with randomly connected circuits,” Proceedings of the National Academy of Sciences of the United States of America, vol. 117, no. 40, pp. 25066–25073, 2020.
Maoz O, Tkačik G, Esteki MS, Kiani R, Schneidman E. 2020. Learning probabilistic neural representations with randomly connected circuits. Proceedings of the National Academy of Sciences of the United States of America. 117(40), 25066–25073.
Maoz, Ori, et al. “Learning Probabilistic Neural Representations with Randomly Connected Circuits.” Proceedings of the National Academy of Sciences of the United States of America, vol. 117, no. 40, National Academy of Sciences, 2020, pp. 25066–73, doi:10.1073/pnas.1912804117.
All files available under the following license(s):
Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0):
Main File(s)
File Name
Access Level
OA Open Access
Date Uploaded
2020-10-27
MD5 Checksum
c6a24fdecf3f28faf447078e7a274a88


Export

Marked Publications

Open Data IST Research Explorer

Sources

PMID: 32948691
PubMed | Europe PMC

Search this title in

Google Scholar