Trust the Batch, On- or Off-Policy: Adaptive Policy Optimization for RL Post-Training

• URL: http://arxiv.org/abs/2605.12380v1
• Date: Tue, 12 May 2026 16:44:47 GMT
• Title: Trust the Batch, On- or Off-Policy: Adaptive Policy Optimization for RL Post-Training
• Abstract summary: Reinforcement learning is structurally harder than supervised learning because the policy changes the data distribution it learns from.<n>We present a simple yet effective batch-adaptive objective that replaces fixed clipping with the normalized effective sample size of the policy ratios.
• Score: 50.86545293331458
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reinforcement learning is structurally harder than supervised learning because the policy changes the data distribution it learns from. The resulting fragility is especially visible in large-model training, where the training and rollout systems differ in numerical precision, sampling, and other implementation details. Existing methods manage this fragility by adding hyper-parameters to the training objective, which makes the algorithm more sensitive to its configuration and requires retuning whenever the task, model scale, or distribution mismatch changes. This fragility traces to two concerns that current objectives entangle through hyper-parameters set before training begins: a trust-region concern, that updates should not move the policy too far from its current value, and an off-policy concern, that data from older or different behavior policies should influence the update only to the extent that it remains reliable. Neither concern is a constant to set in advance, and their severity is reflected in the policy-ratio distribution of the current batch. We present a simple yet effective batch-adaptive objective that replaces fixed clipping with the normalized effective sample size of the policy ratios. The same statistic caps the score-function weight and sets the strength of an off-policy regularizer, so the update stays close to the usual on-policy score-function update when ratios are nearly uniform, and tightens automatically when stale or mismatched data cause ratio concentration, while retaining a nonzero learning signal on high-ratio tokens. Experiments across a wide range of settings show that our method matches or exceeds tuned baselines, introducing no new objective hyper-parameters and removing several existing ones. The code is available at https://github.com/FeynRL-project/FeynRL.

Related papers

• RL Forgets! Towards Continual Policy Optimization [57.4336338996653]
  Continual post-training is becoming a central paradigm for adapting vision-language models to evolving tasks.<n>Recent work has increasingly favored reinforcement learning over supervised fine-tuning, driven by the belief that reinforcement learning is inherently less prone to forgetting.<n>We propose Continual Policy Optimization (CPO), a replay-free framework grounded in a prior-task behavioral KL objective.
  arXiv  Detail & Related papers  (2026-07-05T15:46:11Z)
• Off-Policy Learning to Reason Works Because It Is More Pessimistic Than You Think [6.011610712471053]
  We provide an intuitive construction of off-policy objectives that include successful off-policy objectives.<n>We show that their effectiveness can be understood through implicit pessimism.<n>We then propose a principled modification that stabilize this induced distribution and improve off-policy learning.
  arXiv  Detail & Related papers  (2026-05-27T08:33:35Z)
• Mind the Gap: Data Rewriting for Stable Off-Policy Supervised Fine-Tuning [33.899779762210976]
  Supervised fine-tuning (SFT) of large language models can be viewed as an off-policy learning problem.<n>Existing methods mitigate this issue with KL penalties or clipping, which passively updates rather than actively reducing the gap.<n>We propose a simple yet effective data rewriting framework that proactively shrinks the policy gap before training.
  arXiv  Detail & Related papers  (2025-09-18T17:02:30Z)
• Relative Entropy Pathwise Policy Optimization [66.03329137921949]
  We present an on-policy algorithm that trains Q-value models purely from on-policy trajectories.<n>We show how to combine policies for exploration with constrained updates for stable training, and evaluate important architectural components that stabilize value function learning.
  arXiv  Detail & Related papers  (2025-07-15T06:24:07Z)
• Off-Policy Evaluation for Large Action Spaces via Policy Convolution [60.6953713877886]
  Policy Convolution family of estimators uses latent structure within actions to strategically convolve the logging and target policies. Experiments on synthetic and benchmark datasets demonstrate remarkable mean squared error (MSE) improvements when using PC.
  arXiv  Detail & Related papers  (2023-10-24T01:00:01Z)
• Normality-Guided Distributional Reinforcement Learning for Continuous Control [13.818149654692863]
  Learning a predictive model of the mean return, or value function, plays a critical role in many reinforcement learning algorithms.<n>We study the value distribution in several continuous control tasks and find that the learned value distribution is empirically quite close to normal.<n>We propose a policy update strategy based on the correctness as measured by structural characteristics of the value distribution not present in the standard value function.
  arXiv  Detail & Related papers  (2022-08-28T02:52:10Z)
• Offline Reinforcement Learning with Implicit Q-Learning [85.62618088890787]
  Current offline reinforcement learning methods need to query the value of unseen actions during training to improve the policy. We propose an offline RL method that never needs to evaluate actions outside of the dataset. This method enables the learned policy to improve substantially over the best behavior in the data through generalization.
  arXiv  Detail & Related papers  (2021-10-12T17:05:05Z)
• BRAC+: Improved Behavior Regularized Actor Critic for Offline Reinforcement Learning [14.432131909590824]
  Offline Reinforcement Learning aims to train effective policies using previously collected datasets. Standard off-policy RL algorithms are prone to overestimations of the values of out-of-distribution (less explored) actions. We improve the behavior regularized offline reinforcement learning and propose BRAC+.
  arXiv  Detail & Related papers  (2021-10-02T23:55:49Z)
• MUSBO: Model-based Uncertainty Regularized and Sample Efficient Batch Optimization for Deployment Constrained Reinforcement Learning [108.79676336281211]
  Continuous deployment of new policies for data collection and online learning is either cost ineffective or impractical. We propose a new algorithmic learning framework called Model-based Uncertainty regularized and Sample Efficient Batch Optimization. Our framework discovers novel and high quality samples for each deployment to enable efficient data collection.
  arXiv  Detail & Related papers  (2021-02-23T01:30:55Z)
• Relative Importance Sampling for off-Policy Actor-Critic in Deep Reinforcement Learning [32.66049977978746]
  Off-policy learning exhibits greater instability when compared to on-policy learning in reinforcement learning (RL)<n>We propose a smooth form of importance sampling, specifically relative importance sampling (RIS), which mitigates variance and stabilizes learning.<n>Our methods performed better than or equal to several state-of-the-art RL benchmarks on OpenAI Gym challenges and synthetic datasets.
  arXiv  Detail & Related papers  (2018-10-30T07:41:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.