Generalisation in RL Reading List

Curated by Mouhssine Rifaki | Stanford Electrical Engineering | Last updated August 2026

Papers studying generalisation in reinforcement learning, including evaluation protocols, benchmarks, and algorithmic advances.

Generalisation in RL: 10 key papers

1. A Survey of Zero-shot Generalisation in Deep Reinforcement Learning
   Kirk et al. arXiv 2021.
2. Quantifying Generalization in Reinforcement Learning
   Cobbe et al. arXiv 2018.
3. Leveraging Procedural Generation to Benchmark Reinforcement Learning
   Cobbe et al. arXiv 2019.
4. Assessing Generalization in Deep Reinforcement Learning
   Packer et al. arXiv 2018.
5. Observational Overfitting in Reinforcement Learning
   Song et al. arXiv 2019.
6. Network Randomization: A Simple Technique for Generalization in Deep Reinforcement Learning
   Lee et al. arXiv 2019.
7. Decoupling Value and Policy for Generalization in Reinforcement Learning
   Raileanu and Fergus. arXiv 2021.
8. Why Generalization in RL is Difficult: Epistemic POMDPs and Implicit Partial Observability
   Ghosh et al. arXiv 2021.
9. Investigating Generalisation in Continuous Deep Reinforcement Learning
   Zhao et al. arXiv 2019.
10. A Study on Overfitting in Deep Reinforcement Learning
    Zhang et al. arXiv 2018.

Related reading lists

• Zero-Shot RL & Generalization
• Off-Policy Evaluation & Offline Evaluation
• Retrieval-Augmented Generation
• Sim-to-Real Transfer
• Causal Inference & RL
• Safe Reinforcement Learning & Constrained MDPs
• Transformers for Sequential Decision Making
• Instruction Following & Finetuning