conference paper
Liquid time-constant networks
Technical Tracks
published
yes
Ramin
Hasani
author
Mathias
Lechner
author 3DC22916-F248-11E8-B48F-1D18A9856A87
Alexander
Amini
author
Daniela
Rus
author
Radu
Grosu
author
GradSch
department
ToHe
department
AAAI: Association for the Advancement of Artificial Intelligence
The Wittgenstein Prize
project
We introduce a new class of time-continuous recurrent neural network models. Instead of declaring a learning system’s dynamics by implicit nonlinearities, we construct networks of linear first-order dynamical systems modulated via nonlinear interlinked gates. The resulting models represent dynamical systems with varying (i.e., liquid) time-constants coupled to their hidden state, with outputs being computed by numerical differential equation solvers. These neural networks exhibit stable and bounded behavior, yield superior expressivity within the family of neural ordinary differential equations, and give rise to improved performance on time-series prediction tasks. To demonstrate these properties, we first take a theoretical approach to find bounds over their dynamics, and compute their expressive power by the trajectory length measure in a latent trajectory space. We then conduct a series of time-series prediction experiments to manifest the approximation capability of Liquid Time-Constant Networks (LTCs) compared to classical and modern RNNs.
https://research-explorer.app.ist.ac.at/download/10671/10678/16936-Article Text-20430-1-2-20210518 (1).pdf
application/pdfno
AAAI Press2021Virtual
eng
Proceedings of the AAAI Conference on Artificial Intelligence
2159-5399
2374-3468
978-1-57735-866-4
2006.04439
3597657-7666
Hasani, R., Lechner, M., Amini, A., Rus, D., & Grosu, R. (2021). Liquid time-constant networks. In <i>Proceedings of the AAAI Conference on Artificial Intelligence</i> (Vol. 35, pp. 7657–7666). Virtual: AAAI Press.
R. Hasani, M. Lechner, A. Amini, D. Rus, R. Grosu, in:, Proceedings of the AAAI Conference on Artificial Intelligence, AAAI Press, 2021, pp. 7657–7666.
Hasani, Ramin, Mathias Lechner, Alexander Amini, Daniela Rus, and Radu Grosu. “Liquid Time-Constant Networks.” In <i>Proceedings of the AAAI Conference on Artificial Intelligence</i>, 35:7657–66. AAAI Press, 2021.
Hasani R, Lechner M, Amini A, Rus D, Grosu R. Liquid time-constant networks. In: <i>Proceedings of the AAAI Conference on Artificial Intelligence</i>. Vol 35. AAAI Press; 2021:7657-7666.
R. Hasani, M. Lechner, A. Amini, D. Rus, and R. Grosu, “Liquid time-constant networks,” in <i>Proceedings of the AAAI Conference on Artificial Intelligence</i>, Virtual, 2021, vol. 35, no. 9, pp. 7657–7666.
Hasani R, Lechner M, Amini A, Rus D, Grosu R. 2021. Liquid time-constant networks. Proceedings of the AAAI Conference on Artificial Intelligence. AAAI: Association for the Advancement of Artificial Intelligence, Technical Tracks, vol. 35, 7657–7666.
Hasani, Ramin, et al. “Liquid Time-Constant Networks.” <i>Proceedings of the AAAI Conference on Artificial Intelligence</i>, vol. 35, no. 9, AAAI Press, 2021, pp. 7657–66.
106712022-01-25T15:48:36Z2022-05-24T06:36:54Z