mlsanityData to 7 Oct 2026

Safe Reinforcement Learning From Pixels Using a Stochastic Latent Representation

arXiv 2210.01801 ↗First posted 2 Oct 2022cs.LG cs.AIcited by 11

Not on the map. Why?

Abstract (arXiv)

We address the problem of safe reinforcement learning from pixel observations. Inherent challenges in such settings are (1) a trade-off between reward optimization and adhering to safety constraints, (2) partial observability, and (3) high-dimensional observations. We formalize the problem in a constrained, partially observable Markov decision process framework, where an agent obtains distinct reward and safety signals. To address the curse of dimensionality, we employ a novel safety critic using the stochastic latent actor-critic (SLAC) approach. The latent variable model predicts rewards and safety violations, and we use the safety critic to train safe policies. Using well-known benchmark environments, we demonstrate competitive performance over existing approaches with respects to computational requirements, final reward return, and satisfying the safety constraints.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 19 of its 40 references that are in the map. First posted on arXiv 2 Oct 2022.

Builds on

19 of 40 references.

References · 19

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,635
  3. 2013Auto-Encoding Variational Bayescited by 3,980
  4. 2015Trust Region Policy Optimizationcited by 2,503
  5. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,302
  6. 2016Concrete Problems in AI Safetycited by 1,582
  7. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,221
  8. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,179
  9. 2017Constrained Policy Optimizationcited by 928
  10. 2017Safe Reinforcement Learning via Shieldingcited by 427
  11. 2020Responsive Safety in Reinforcement Learning by PID Lagrangian Methodscited by 302
  12. 2019Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Modelcited by 228
7 more
  1. 2021Maximum Entropy RL (Provably) Solves Some Robust RL Problemscited by 169
  2. 2020Learning to Walk in the Real World with Minimal Human Effortcited by 103
  3. 2021Scalar reward is not enough: A response to Silver, Singh, Precup and Sutton (2021)cited by 55
  4. 2022Constrained Policy Optimization via Bayesian World Modelscited by 55
  5. 2022Safe Reinforcement Learning via Shielding under Partial Observabilitycited by 49
  6. 2021Direct Behavior Specification via Constrained Reinforcement Learningcited by 30
  7. 2021Safe Continuous Control with Constrained Model-Based Policy Optimizationcited by 17

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 1; 2025: 2; 2026: 4.

11 papers.

  1. 2023SafeDreamer: Safe Reinforcement Learning with World Modelscited by 50
  2. 2023State-Wise Safe Reinforcement Learning With Pixel Observationscited by 9
  3. 2025PIGDreamer: Privileged Information Guided World Models for Safe Partially Observable Reinforcement Learningcited by 7
  4. 2023Explainability Techniques for Chemical Language Modelscited by 6
  5. 2024FOSP: Fine-tuning Offline Safe Policy through World Modelscited by 3
  6. 2026Formally Verifying and Explaining Sepsis Treatment Policies with COOL-MCcited by 3
  7. 2023Using Cooperative Game Theory to Prune Neural Networkscited by 1
  8. 2025Enhance Exploration in Safe Reinforcement Learning with Contrastive Representation Learningcited by 0
  9. 2026Nightmare Dreamer: Dreaming About Unsafe States And Planning Aheadcited by 0
  10. 2026COOL-MC: Verifying and Explaining RL Policies for Platelet Inventory Managementcited by 0
  11. 2026Imagine to Ensure Safety in Hierarchical Reinforcement Learningcited by 0