DR-APO: Decoupled-Representation Advantage-Guided Policy Optimization from Continuous Control to Embodied Navigation
cs.LG, cs.RO
Submitted: 2026-10-12
Updated: 2026-10-12
Comments: 9 pages, 4 figures,
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Flow Matching for Generative Modeling
- Proximal Policy Optimization Algorithms
- Diffusion Policy Policy Optimization
- Efficient Online Reinforcement Learning for Diffusion Policy
- ReinFlow: Fine-tuning Flow Matching Policy with Online Reinforcement Learning
- $\pi_\texttt{RL}$: Online RL Fine-tuning for Flow-based Vision-Language-Action Models
- Diffusion-based Reinforcement Learning via Q-weighted Variational Policy Optimization
- Reverse Flow Matching: A Unified Framework for Online Reinforcement Learning with Diffusion and Flow Policies
- Continuous control with deep reinforcement learning
- Addressing Function Approximation Error in Actor-Critic Methods
- Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
- Maximum a Posteriori Policy Optimisation
- Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
- AWAC: Accelerating Online Reinforcement Learning with Offline Datasets
- Critic Regularized Regression
- Planning with Diffusion for Flexible Behavior Synthesis
- Is Conditional Generative Modeling all you need for Decision-Making?
- Diffusion Policies as an Expressive Policy Class for Offline Reinforcement Learning
- Flow Matching Policy Gradients
- GenPO: Generative Diffusion Models Meet On-Policy 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