mlsanityData to 7 Oct 2026

Deluca -- A Differentiable Control Library: Environments, Methods, and Benchmarking

arXiv 2102.09968 ↗First posted 19 Feb 2021cs.RO cs.LGcited by 10

Not on the map. Why?

Abstract (arXiv)

We present an open-source library of natively differentiable physics and robotics environments, accompanied by gradient-based control methods and a benchmark-ing suite. The introduced environments allow auto-differentiation through the simulation dynamics, and thereby permit fast training of controllers. The library features several popular environments, including classical control settings from OpenAI Gym. We also provide a novel differentiable environment, based on deep neural networks, that simulates medical ventilation. We give several use-cases of new scientific results obtained using the library. This includes a medical ventilator simulator and controller, an adaptive control method for time-varying linear dynamical systems, and new gradient-based methods for control of linear dynamical systems with adversarial perturbations.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 11 of its 43 references that are in the map. First posted on arXiv 19 Feb 2021.

Builds on

11 of 43 references.

References · 11

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2015Trust Region Policy Optimizationcited by 2,503
  3. 2018DeepMind Control Suitecited by 935
  4. 2019Introduction to Online Convex Optimizationcited by 793
  5. 2019DiffTaichi: Differentiable Programming for Physical Simulationcited by 248
  6. 2016A Differentiable Physics Engine for Deep Learning in Roboticscited by 107
  7. 2018ChainQueen: A Real-Time Differentiable Physical Simulator for Soft Roboticscited by 98
  8. 2019Online Control with Adversarial Disturbancescited by 92
  9. 2019The Nonstochastic Control Problemcited by 52
  10. 2020Black-Box Control for Linear Dynamical Systemscited by 43
  11. 2020Adaptive Regret for Control of Time-Varying Dynamicscited by 38

Built on by

Cited by year

Citing papers per year: 2021: 3; 2022: 3; 2023: 4.

10 papers.

  1. 2021Brax -- A Differentiable Physics Engine for Large Scale Rigid Body Simulationcited by 375
  2. 2022Do Differentiable Simulators Give Better Policy Gradients?cited by 83
  3. 2021Gradients are Not All You Needcited by 82
  4. 2023Physics-Informed Machine Learning for Modeling and Control of Dynamical Systemscited by 28
  5. 2022Training Efficient Controllers via Analytic Policy Gradientcited by 16
  6. 2021A Regret Minimization Approach to Iterative Learning Controlcited by 8
  7. 2023Optimal Rates for Bandit Nonstochastic Controlcited by 7
  8. 2022A Regret Minimization Approach to Multi-Agent Controlcited by 4
  9. 2023Controlgym: Large-Scale Control Environments for Benchmarking Reinforcement Learning Algorithmscited by 4
  10. 2023Online Nonstochastic Model-Free Reinforcement Learningcited by 2