Investigating Memory in Model-Free RL with POPGym Arcade
cs.LG, cs.AI, cs.RO
Submitted: 2025-03-03
Updated: 2026-08-27
Comments: Appear at ICML 2026 as a Spotlight paper
Code: https://github.com/bolt-research/popgym-arcade
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- DeepMind Lab
- Simplifying Deep Temporal Difference Learning
- Jumanji: a Diverse Suite of Scalable Reinforcement Learning Environments in JAX
- Podracer architectures for scalable Reinforcement Learning
- ViZDoom: A Doom-based AI Research Platform for Visual Reinforcement Learning
- Kinetix: Investigating the Training of General Agents through Open-Ended Physics-Based Control Tasks
- NAVIX: Scaling MiniGrid Environments with JAX
- MDP Playground: An Analysis and Debug Testbed for Reinforcement Learning
- Proximal Policy Optimization Algorithms
- The Distracting Control Suite -- A Challenging Benchmark for Reinforcement Learning from Pixels
- RoFormer: Enhanced Transformer with Rotary Position Embedding
- MinAtar: An Atari-Inspired Testbed for Thorough and Reproducible Reinforcement Learning Experiments
- Benchmarking Partial Observability in Reinforcement Learning with a Suite of Memory-Improvable Domains
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks