RAMAC: Multimodal Risk-Aware Offline Reinforcement Learning and the Role of Behavior Regularization

arXiv:2510.02695 · cs.LG, cs.AI · Submitted 2025-10-03 · Read on arXiv

cs.LG, cs.AI

Submitted: 2025-10-03

Updated: 2026-06-11

Comments: ICML 2026

Journal ref: Proceedings of Machine Learning Research, 306:32295-32324, 2026 (ICML 2026)

Project page: https://kaifukazawa.github.io/ramac-project

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

Terminology

• Actor — 3× in this paper · explained in 3 episodes
• Reinforcement Learning — 2× in this paper · explained in 4 episodes
• Conditional Value-at-Risk — 1× in this paper
• Regularization — 1× in this paper · explained in 3 episodes
• Reinforcement Learning (RL) — 1× in this paper · explained in 25 episodes
• Risk-Aware Multimodal Actor-Critic — 1× in this paper

Sources

• D4RL: Datasets for Deep Data-Driven Reinforcement Learning
• IDQL: Implicit Q-Learning as an Actor-Critic Method with Diffusion Policies
• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
• Flow Matching for Generative Modeling
• AWAC: Accelerating Online Reinforcement Learning with Offline Datasets
• Behavior Regularized Offline Reinforcement Learning
• ManiFlow: A General Robot Manipulation Policy via Consistency Flow Training

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