mlsanityData to 7 Oct 2026

Physics-Regulated Deep Reinforcement Learning: Invariant Embeddings

arXiv 2305.16614 ↗First posted 26 May 2023cs.AI cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

This paper proposes the Phy-DRL: a physics-regulated deep reinforcement learning (DRL) framework for safety-critical autonomous systems. The Phy-DRL has three distinguished invariant-embedding designs: i) residual action policy (i.e., integrating data-driven-DRL action policy and physics-model-based action policy), ii) automatically constructed safety-embedded reward, and iii) physics-model-guided neural network (NN) editing, including link editing and activation editing. Theoretically, the Phy-DRL exhibits 1) a mathematically provable safety guarantee and 2) strict compliance of critic and actor networks with physics knowledge about the action-value function and action policy. Finally, we evaluate the Phy-DRL on a cart-pole system and a quadruped robot. The experiments validate our theoretical results and demonstrate that Phy-DRL features guaranteed safety compared to purely data-driven DRL and solely model-based design while offering remarkably fewer learning parameters and fast training towards safety guarantee.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 34 of its 54 references that are in the map. First posted on arXiv 26 May 2023.

Builds on

34 of 54 references.

References · 34

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  3. 2015Continuous control with deep reinforcement learningcited by 3,279
  4. 2019When to Trust Your Model: Model-Based Policy Optimizationcited by 753
  5. 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 739
  6. 2016TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systemscited by 660
  7. 2016CVXPY: A Python-Embedded Modeling Language for Convex Optimizationcited by 633
  8. 2016Geometric deep learning on graphs and manifolds using mixture model CNNscited by 468
  9. 2020Lagrangian Neural Networkscited by 388
  10. 2017Safe Model-based Reinforcement Learning with Stability Guaranteescited by 347
  11. 2017Adversarial Attacks on Neural Network Policiescited by 314
  12. 2021Physics-informed neural networks with hard constraints for inverse designcited by 277
22 more
  1. 2018Learning to Drive in a Daycited by 266
  2. 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 256
  3. 2018Residual Reinforcement Learning for Robot Controlcited by 242
  4. 2019Towards Physics-informed Deep Learning for Turbulent Flow Predictioncited by 203
  5. 2020Integrating Scientific Knowledge with Machine Learning for Engineering and Environmental Systemscited by 160
  6. 2020Incorporating Symmetry into Deep Dynamics Models for Improved Generalizationcited by 123
  7. 2018Non-asymptotic Identification of LTI Systems from a Single Trajectorycited by 77
  8. 2021Learning Vision-Guided Quadrupedal Locomotion End-to-End with Cross-Modal Transformerscited by 71
  9. 2019Learning Compositional Koopman Operators for Model-Based Controlcited by 67
  10. 2020Physics-Guided Machine Learning for Scientific Discovery: An Application in Simulating Lake Temperature Profilescited by 59
  11. 2021Physics-Guided Deep Learning for Dynamical Systems: A Surveycited by 43
  12. 2019Control Regularization for Reduced Variance Reinforcement Learningcited by 34
  13. 2021Stabilizing Neural Control Using Self-Learned Almost Lyapunov Criticscited by 33
  14. 2020Theory-guided hard constraint projection (HCP): a knowledge-based data-driven scientific machine learning methodcited by 29
  15. 2020Learning a Contact-Adaptive Controller for Robust, Efficient Legged Locomotioncited by 22
  16. 2022Lyapunov Design for Robust and Efficient Robotic Reinforcement Learningcited by 20
  17. 2021Bayesian Controller Fusion: Leveraging Control Priors in Deep Reinforcement Learning for Roboticscited by 18
  18. 2020Isometric Transformation Invariant and Equivariant Graph Convolutional Networkscited by 17
  19. 2023Stable and Safe Reinforcement Learning via a Barrier-Lyapunov Actor-Critic Approachcited by 11
  20. 2020SL1-Simplex: Safe Velocity Regulation of Self-Driving Vehicles in Dynamic and Unforeseen Environmentscited by 7
  21. 2022Equipping Black-Box Policies with Model-Based Advice for Stable Nonlinear Controlcited by 5
  22. 2020Social System Inference from Noisy Observationscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 1; 2026: 4.

8 papers.

  1. 2024Stable and Safe Human-aligned Reinforcement Learning through Neural Ordinary Differential Equationscited by 3
  2. 2024Simplex-enabled Safe Continual Learning Machinecited by 2
  3. 2026TRIDENT: Breaking the Hybrid-Safety-Physics Coupling for Provably Safe Multi-Agent Reinforcement Learningcited by 2
  4. 2026Efficient Real-World Autonomous Racing via Attenuated Residual Policy Optimizationcited by 1
  5. 2024Physics-model-guided Worst-case Sampling for Safe Reinforcement Learningcited by 0
  6. 2025Real-DRL: Teach and Learn in Realitycited by 0
  7. 2026Safe Online Learning via Smooth Safety-Structured Policy Compositioncited by 0
  8. 2026Safe and Robust Neural Policy Learning with Statistical Verification for Sim-to-Real Deployment in Roboticscited by 0