Learning Integrable Dynamics with Action-Angle Networks
Machine learning has become increasingly popular for efficiently modelling the dynamics of complex physical systems, demonstrating a capability to learn effective models for dynamics which ignore redundant degrees of freedom. Learned simulators typically predict the evolution of the system in a step-by-step manner with numerical integration techniques. However, such models often suffer from instability over long roll-outs due to the accumulation of both estimation and integration error at each prediction step. Here, we propose an alternative construction for learned physical simulators that are inspired by the concept of action-angle coordinates from classical mechanics for describing integrable systems. We propose Action-Angle Networks, which learn a nonlinear transformation from input coordinates to the action-angle space, where evolution of the system is linear. Unlike traditional learned simulators, Action-Angle Networks do not employ any higher-order numerical integration methods, making them extremely efficient at modelling the dynamics of integrable physical systems.
Code (1)
Tasks
Numerical IntegrationSimilar Papers 제목 키워드 기반
Machine learning prediction for mean motion resonance behaviour -- The planar case
Most recently, machine learning has been used to study the dynamics of integrable Hamiltonian systems and the chaotic 3-body problem. In this work, we consider an intermediate case of regular motion in a non-integrable s…
BIG-bench Machine LearningNumerical IntegrationAbsence of Closed-Form Descriptions for Gradient Flow in Two-Layer Narrow Networks
In the field of machine learning, comprehending the intricate training dynamics of neural networks poses a significant challenge. This paper explores the training dynamics of neural networks, particularly whether these d…
FormMetric-Gradient Projection for Stable Multi-Agent Policy Learning
General-sum multi-agent learning is often governed by a stacked update field in which each agent's policy update changes the optimization landscape faced by the others. This coupling can entangle an integrable component …
Training Integrable Parameterizations of Deep Neural Networks in the Infinite-Width Limit
To theoretically understand the behavior of trained deep neural networks, it is necessary to study the dynamics induced by gradient methods from a random initialization. However, the nonlinear and compositional structure…
image-classificationImage ClassificationCochain Perspectives on Temporal-Difference Signals for Learning Beyond Markov Dynamics
Non-Markovian dynamics are commonly found in real-world environments due to long-range dependencies, partial observability, and memory effects. The Bellman equation that is the central pillar of Reinforcement learning (R…
Reinforcement Learning