Learning locally interacting discrete dynamical systems: Towards data-efficient and scalable prediction

Beomseok Kang, Harshit Kumar, Minah Lee, Biswadeep Chakraborty, Saibal Mukhopadhyay
Proceedings of the 6th Annual Learning for Dynamics & Control Conference, PMLR 242:1357-1369, 2024.

Abstract

Locally interacting dynamical systems, such as epidemic spread, rumor propagation through crowd, and forest fire, exhibit complex global dynamics originated from local, relatively simple, and often stochastic interactions between dynamic elements. Their temporal evolution is often driven by transitions between a finite number of discrete states. Despite significant advancements in predictive modeling through deep learning, such interactions among many elements have rarely explored as a specific domain for predictive modeling. We present Attentive Recurrent Neural Cellular Automata (AR-NCA), to effectively discover unknown local state transition rules by associating the temporal information between neighboring cells in a permutation-invariant manner. AR-NCA exhibits the superior generalizability across various system configurations (i.e., spatial distribution of states), data efficiency and robustness in extremely data-limited scenarios even in the presence of stochastic interactions, and scalability through spatial dimension-independent prediction.

Cite this Paper

BibTeX
@InProceedings{pmlr-v242-kang24a, title = {Learning locally interacting discrete dynamical systems: {T}owards data-efficient and scalable prediction}, author = {Kang, Beomseok and Kumar, Harshit and Lee, Minah and Chakraborty, Biswadeep and Mukhopadhyay, Saibal}, booktitle = {Proceedings of the 6th Annual Learning for Dynamics & Control Conference}, pages = {1357--1369}, year = {2024}, editor = {Abate, Alessandro and Cannon, Mark and Margellos, Kostas and Papachristodoulou, Antonis}, volume = {242}, series = {Proceedings of Machine Learning Research}, month = {15--17 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v242/kang24a/kang24a.pdf}, url = {https://proceedings.mlr.press/v242/kang24a.html}, abstract = {Locally interacting dynamical systems, such as epidemic spread, rumor propagation through crowd, and forest fire, exhibit complex global dynamics originated from local, relatively simple, and often stochastic interactions between dynamic elements. Their temporal evolution is often driven by transitions between a finite number of discrete states. Despite significant advancements in predictive modeling through deep learning, such interactions among many elements have rarely explored as a specific domain for predictive modeling. We present Attentive Recurrent Neural Cellular Automata (AR-NCA), to effectively discover unknown local state transition rules by associating the temporal information between neighboring cells in a permutation-invariant manner. AR-NCA exhibits the superior generalizability across various system configurations (i.e., spatial distribution of states), data efficiency and robustness in extremely data-limited scenarios even in the presence of stochastic interactions, and scalability through spatial dimension-independent prediction.} }
Endnote
%0 Conference Paper %T Learning locally interacting discrete dynamical systems: Towards data-efficient and scalable prediction %A Beomseok Kang %A Harshit Kumar %A Minah Lee %A Biswadeep Chakraborty %A Saibal Mukhopadhyay %B Proceedings of the 6th Annual Learning for Dynamics & Control Conference %C Proceedings of Machine Learning Research %D 2024 %E Alessandro Abate %E Mark Cannon %E Kostas Margellos %E Antonis Papachristodoulou %F pmlr-v242-kang24a %I PMLR %P 1357--1369 %U https://proceedings.mlr.press/v242/kang24a.html %V 242 %X Locally interacting dynamical systems, such as epidemic spread, rumor propagation through crowd, and forest fire, exhibit complex global dynamics originated from local, relatively simple, and often stochastic interactions between dynamic elements. Their temporal evolution is often driven by transitions between a finite number of discrete states. Despite significant advancements in predictive modeling through deep learning, such interactions among many elements have rarely explored as a specific domain for predictive modeling. We present Attentive Recurrent Neural Cellular Automata (AR-NCA), to effectively discover unknown local state transition rules by associating the temporal information between neighboring cells in a permutation-invariant manner. AR-NCA exhibits the superior generalizability across various system configurations (i.e., spatial distribution of states), data efficiency and robustness in extremely data-limited scenarios even in the presence of stochastic interactions, and scalability through spatial dimension-independent prediction.
APA
Kang, B., Kumar, H., Lee, M., Chakraborty, B. & Mukhopadhyay, S.. (2024). Learning locally interacting discrete dynamical systems: Towards data-efficient and scalable prediction. Proceedings of the 6th Annual Learning for Dynamics & Control Conference, in Proceedings of Machine Learning Research 242:1357-1369 Available from https://proceedings.mlr.press/v242/kang24a.html.

Related Material