Horizon Adaptive Offline Policy Learning via Value Stitching

arXiv:2606.21136 · cs.LG, cs.RO · Submitted 2026-06-19 · Read on arXiv

cs.LG, cs.RO

Submitted: 2026-06-19

Updated: 2026-10-11

Comments: NeurIPS 2026

Code: https://github.com/Whiterrrrr/value-stitching

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Reinforcement Learning (RL) — 1× in this paper · explained in 25 episodes

Sources

• Chunking the Critic: A Transformer-based Soft Actor-Critic with N-Step Returns
• Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
• Q-learning with Adjoint Matching

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