From Static Policies to Adaptive Priors in Offline Reinforcement Learning

arXiv:2609.35880 · cs.LG · Submitted 2026-09-26 · Read on arXiv

cs.LG

Submitted: 2026-09-26

Updated: 2026-09-26

Terminology

• Reinforcement Learning — 1× in this paper · explained in 4 episodes

Sources

• Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting
• RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning
• Why AI systems don't learn and what to do about it: Lessons on autonomous learning from cognitive science
• D4RL: Datasets for Deep Data-Driven Reinforcement Learning
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• Contextual Markov Decision Processes
• Scaling Laws for Neural Language Models
• Plasticity Loss in Deep Reinforcement Learning: A Survey
• Improving Interactive In-Context Learning from Natural Language Feedback
• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
• Uncertainty-Aware Robotic World Model Makes Offline Model-Based Reinforcement Learning Work on Real Robots
• The Three Regimes of Offline-to-Online Reinforcement Learning
• Finetuning from Offline Reinforcement Learning: Challenges, Trade-offs and Practical Solutions
• A Survey of In-Context Reinforcement Learning
• AWAC: Accelerating Online Reinforcement Learning with Offline Datasets
• Assessing Generalization in Deep Reinforcement Learning
• Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
• Posterior Behavioral Cloning: Pretraining BC Policies for Efficient RL Finetuning
• Behavior Regularized Offline Reinforcement Learning
• Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning

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