Efficient Online Reinforcement Learning for Diffusion Policy
read the original abstract
Diffusion policies have achieved superior performance in imitation learning and offline reinforcement learning (RL) due to their rich expressiveness. However, the conventional diffusion training procedure requires samples from target distribution, which is impossible in online RL since we cannot sample from the optimal policy. Backpropagating policy gradient through the diffusion process incurs huge computational costs and instability, thus being expensive and not scalable. To enable efficient training of diffusion policies in online RL, we generalize the conventional denoising score matching by reweighting the loss function. The resulting Reweighted Score Matching (RSM) preserves the optimal solution and low computational cost of denoising score matching, while eliminating the need to sample from the target distribution and allowing learning to optimize value functions. We introduce two tractable reweighted loss functions to solve two commonly used policy optimization problems, policy mirror descent and max-entropy policy, resulting in two practical algorithms named Diffusion Policy Mirror Descent (DPMD) and Soft Diffusion Actor-Critic (SDAC). We conducted comprehensive comparisons on MuJoCo benchmarks. The empirical results show that the proposed algorithms outperform recent diffusion-policy online RLs on most tasks, and the DPMD improves more than 120% over soft actor-critic on Humanoid and Ant.
This paper has not been read by Pith yet.
Forward citations
Cited by 14 Pith papers
-
Reinforcement Learning for Flow-Matching Policies with Density Transport
RLDT fine-tunes pretrained flow-matching policies for continuous control by aligning them to a max-entropy RL transport field constructed via SVGD, using expected-target estimation for stable multi-step updates.
-
Block-R1: Rethinking the Role of Block Size in Multi-domain Reinforcement Learning for Diffusion Large Language Models
Introduces Block-R1 benchmark, Block-R1-41K dataset, and a conflict score to handle domain-specific optimal block sizes in RL post-training of diffusion LLMs.
-
Block-R1: Rethinking the Role of Block Size in Multi-domain Reinforcement Learning for Diffusion Large Language Models
Block-R1 formulates domain block size conflicts in multi-domain RL for dLLMs, releases a 41K-sample dataset with per-sample best block sizes and a conflict score, and provides a benchmark plus simple cross-domain trai...
-
Advantage-Guided Diffusion for Model-Based Reinforcement Learning
Advantage-guided diffusion (SAG and EAG) steers sampling in diffusion world models to higher-advantage trajectories, enabling policy improvement and better sample efficiency on MuJoCo tasks.
-
MODIP: Efficient Model-Based Optimization for Diffusion Policies
MODIP fine-tunes diffusion policies offline-to-online by training a world model, running MPC with terminal state values inside it to create targets, and using policy-independent TD critics, yielding gains over BC on D...
-
Revisiting Policy Gradients for Restricted Policy Classes: Escaping Myopic Local Optima with $k$-step Policy Gradients
The k-step policy gradient converges exponentially close to the optimal deterministic policy in restricted classes, achieving O(1/T) rates under smoothness assumptions without distribution mismatch factors.
-
Decentralized Diffusion Policy Learning for Enhanced Exploration in Cooperative Multi-agent Reinforcement Learning
Decentralized diffusion policies trained with importance sampling score matching enhance exploration and performance in cooperative MARL over Gaussian policy baselines.
-
Model-Based Proactive Cost Generation for Learning Safe Policies Offline with Limited Violation Data
PROCO generates synthetic unsafe samples via model-based rollouts and LLM-grounded costs to enable safer policy learning from offline datasets containing few or no violations.
-
Fisher Decorator: Refining Flow Policy via a Local Transport Map
Fisher Decorator refines flow policies in offline RL via a local transport map and Fisher-matrix quadratic approximation of the KL constraint, yielding controllable error near the optimum and SOTA benchmark results.
-
What Does Flow Matching Bring To TD Learning?
Flow matching critics outperform monolithic ones in RL by 2x performance and 5x sample efficiency via test-time error recovery through integration and multi-point velocity supervision that preserves feature plasticity.
-
How Does the Lagrangian Guide Safe Reinforcement Learning through Diffusion Models?
ALGD augments the Lagrangian to locally convexify the energy landscape in diffusion models, stabilizing safe RL training and generation without changing optimal policies.
-
Reinforcement Learning with Discrete Diffusion Policies for Combinatorial Action Spaces
A method trains discrete diffusion policies for combinatorial RL by matching to a PMD-regularized target distribution, reporting SOTA performance and sample efficiency on DNA generation, macro-action, and multi-agent ...
-
BORA: Bridging Offline Reinforcement Learning and Online Residual Adaptation for Real-World Dexterous VLA Models
BORA combines offline RL critic training with online chunk-wise residual adaptation to raise average success rates of real-world dexterous VLA policies by 33% and up to 43% on unseen objects across five tasks.
-
D2 Actor Critic: Diffusion Actor Meets Distributional Critic
D2AC combines a diffusion actor with a distributional critic via fused distributional RL and clipped double Q-learning to reach state-of-the-art results on 18 hard control benchmarks including Humanoid, Dog, and Shadow Hand.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.