Dissipative SymODEN: Encoding Hamiltonian Dynamics with Dissipation and Control into Deep Learning

2020/02/20 by Yaofeng Desmond Zhong, Zhong, Yaofeng Desmond, Biswadip Dey +3 · 10 citations
Computer Science · Engineering · Materials Science · Mathematics · Physics and Astronomy · #FOS: Computer and information sciences #FOS: Electrical engineering #Fuel Cells and Related Materials #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning in Materials Science #Model Reduction and Neural Networks #Systems and Control (eess.SY) #cs.LG #cs.SY #eess.SY #electronic engineering #information engineering #stat.ML

paper · pdf · doi:10.48550/arxiv.2002.08860

Published at ICLR 2020 Workshop on Integration of Deep Neural Models and Differential Equations (DeepDiffEq)

openalex publication_date 2020/02/20 · arxiv created 2020/04/30 · arxiv updated 2020/05/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In this work, we introduce Dissipative SymODEN, a deep learning architecture which can infer the dynamics of a physical system with dissipation from observed state trajectories. To improve prediction accuracy while reducing network size, Dissipative SymODEN encodes the port-Hamiltonian dynamics with energy dissipation and external input into the design of its computation graph and learns the dynamics in a structured way. The learned model, by revealing key aspects of the system, such as the inertia, dissipation, and potential energy, paves the way for energy-based controllers.

Citations

• Adam: A Method for Stochastic Optimization
• Identification and control of dynamical systems using neural networks
• Continuous control with deep reinforcement learning
• A General Framework for Structured Learning of Mechanical Systems
• Hamiltonian Neural Networks

Cited by

• Control-Oriented System Identification: Classical, Learning, and Physics-Informed Approaches
• Learning Hamiltonian Dynamics at Scale: A Differential-Geometric Approach
• Physically Plausible Multi-System Trajectory Generation and Symmetry Discovery
• Learning Generalized Hamiltonian Dynamics with Stability from Noisy Trajectory Data
• Data-driven particle dynamics: Structure-preserving coarse-graining for emergent behavior in non-equilibrium systems
• Can we learn gradients by Hamiltonian Neural Networks?
• Reversible and irreversible bracket-based dynamics for deep graph neural networks
• Geometric Contact Flows: Contactomorphisms for Dynamics and Control
• CoSynFlow: Conformal Symplectic Neural Flows for Cross-System Prediction of Dissipative Hamiltonian Dynamics
• Learning mechanical systems from real-world data using discrete forced Lagrangian dynamics

Related

• Learning Deep Dissipative Dynamics
• Dissipative Hamiltonian Neural Networks: Learning Dissipative and Conservative Dynamics Separately
• Learning Neural Hamiltonian Dynamics: A Methodological Overview
• Stable Port-Hamiltonian Neural Networks
• Identifiable learning of dissipative dynamics