Beyond State-Wise Mirror Descent: Offline Policy Optimization with Parametric Policies
cs.LG, cs.AI
Submitted: 2026-02-27
Updated: 2026-10-09
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Low-Rank MDPs with Continuous Action Spaces
- Offline Reinforcement Learning in Large State Spaces: Algorithms and Guarantees
- Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
- Off-Policy Policy Gradient with State Distribution Correction
- Offline Reinforcement Learning with Additional Covering Distributions
- Playing Atari with Deep Reinforcement Learning
- Online Learning: A Modern Introduction Using Convex Optimization
- Distributionally Robust Optimization: A Review
- Diffusion Policy Policy Optimization
- The Pitfalls of Imitation Learning when Actions are Continuous
- Linear Convergence of Natural Policy Gradient Methods with Log-Linear Policies
- Offline Data Enhanced On-Policy Policy Gradient with Provable Guarantees
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