Learning Locally Interacting Discrete Dynamical Systems: Towards Data-Efficient and Scalable Prediction
This addresses the challenge of predictive modeling for systems with local stochastic interactions, which is incremental as it applies deep learning to a rarely explored domain.
The paper tackles the problem of predicting complex global dynamics in locally interacting discrete dynamical systems, such as epidemic spread, by developing AR-NCA, which achieves superior generalizability, data efficiency, and scalability in data-limited scenarios.
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.