Learning Kernel-Based MDPs from Episodic Preferential Feedback

• URL: http://arxiv.org/abs/2605.23650v2
• Date: Mon, 25 May 2026 02:19:44 GMT
• Title: Learning Kernel-Based MDPs from Episodic Preferential Feedback
• Abstract summary: We present a theoretical study of preference-only learning in episodic kernel MDPs.<n>We develop preference-based value estimation and confidence sets tailored to end-of-episode comparisons.<n>We prove high-probability regret bounds that scale sublinearly in the number of episodes, implying that the value of the learned policy converges to that of the optimal policy.
• Score: 14.488515539527995
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Human feedback often arrives as preferences rather than calibrated numeric rewards, motivating reinforcement learning from preferential feedback, also referred to as reinforcement learning from human feedback (RLHF). We present a rigorous theoretical study of preference-only learning in episodic kernel MDPs. In each episode, the learner deploys two policies from a common start state and receives a single binary label indicating which trajectory is preferred, modeled by a Bradley--Terry--Luce link on the difference of cumulative (unobserved) rewards. Under kernel-based assumptions on the reward and transition functions (one of the most general models amenable to theoretical analysis) we develop preference-based value estimation and confidence sets tailored to end-of-episode comparisons. We prove high-probability regret bounds that scale sublinearly in the number of episodes, implying that the value of the learned policy converges to that of the optimal policy.

Related papers

• EVPO: Explained Variance Policy Optimization for Adaptive Critic Utilization in LLM Post-Training [69.32453275232662]
  We show that a learned critic can inject estimation noise that exceeds the state signal it captures, increasing rather than reducing advantage variance.<n>We propose Explained Variance Policy Optimization (EVPO), which monitors batch-level EV at each training step and adaptively switches between critic-based and batch-mean advantage estimation.
  arXiv  Detail & Related papers  (2026-04-21T14:07:39Z)
• How RLHF Amplifies Sycophancy [23.213056717401418]
  Large language models often exhibit increased sycophantic behavior after preference-based post-training.<n>We identify an explicit amplification mechanism that causally links optimization against a learned reward to bias in the human preference data used for alignment.<n>We propose a training-time intervention designed to neutralize the amplification mechanism itself.
  arXiv  Detail & Related papers  (2026-02-01T03:46:14Z)
• Rectifying Shortcut Behaviors in Preference-based Reward Learning [46.09046818725698]
  In reinforcement learning, preference-based reward models play a central role in aligning large language models to human-aligned behavior.<n>Recent studies show that these models are prone to reward hacking and often fail to generalize well due to over-optimization.<n>We introduce a principled yet flexible approach to mitigate shortcut behaviors in preference-based reward learning.
  arXiv  Detail & Related papers  (2025-10-21T20:08:32Z)
• Best Policy Learning from Trajectory Preference Feedback [11.896067099790962]
  Preference-based Reinforcement Learning (PbRL) offers a more robust alternative.<n>We study the best policy identification problem in PbRL, motivated by post-training optimization of generative models.<n>We propose Posterior Sampling for Preference Learning ($mathsfPSPL$), a novel algorithm inspired by Top-Two Thompson Sampling.
  arXiv  Detail & Related papers  (2025-01-31T03:55:10Z)
• Reward Modeling with Ordinal Feedback: Wisdom of the Crowd [9.034189257088762]
  Learning a reward model (RM) from human preferences has been an important component in aligning large language models. We propose a framework for learning RMs under ordinal feedback. We prove the statistical benefits of ordinal feedback in terms of reducing the Rademacher complexity.
  arXiv  Detail & Related papers  (2024-11-19T20:17:04Z)
• Zeroth-Order Policy Gradient for Reinforcement Learning from Human Feedback without Reward Inference [15.038210624870656]
  Reward inference is a critical intermediate step in the Reinforcement Learning from Human Feedback pipeline.<n>This paper develops two RLHF algorithms without reward inference for general RL problems beyond bandits and deterministic MDP bandit, and general preference models beyond the Bradley-Terry model.
  arXiv  Detail & Related papers  (2024-09-25T22:20:11Z)
• Learning Optimal Deterministic Policies with Stochastic Policy Gradients [62.81324245896716]
  Policy gradient (PG) methods are successful approaches to deal with continuous reinforcement learning (RL) problems. In common practice, convergence (hyper)policies are learned only to deploy their deterministic version. We show how to tune the exploration level used for learning to optimize the trade-off between the sample complexity and the performance of the deployed deterministic policy.
  arXiv  Detail & Related papers  (2024-05-03T16:45:15Z)
• Generalizing Reward Modeling for Out-of-Distribution Preference Learning [3.9160947065896803]
  Preference learning with large language models (LLMs) aims to align the LLMs' generations with human preferences. Due to the difficulty of obtaining human feedback, discretely training reward models for every encountered distribution is challenging. This work addresses OOD PL by optimizing a general reward model through a meta-learning approach.
  arXiv  Detail & Related papers  (2024-02-22T18:20:33Z)
• A Minimaximalist Approach to Reinforcement Learning from Human Feedback [49.45285664482369]
  We present Self-Play Preference Optimization (SPO), an algorithm for reinforcement learning from human feedback. Our approach is minimalist in that it does not require training a reward model nor unstable adversarial training. We demonstrate that on a suite of continuous control tasks, we are able to learn significantly more efficiently than reward-model based approaches.
  arXiv  Detail & Related papers  (2024-01-08T17:55:02Z)
• Provable Offline Preference-Based Reinforcement Learning [95.00042541409901]
  We investigate the problem of offline Preference-based Reinforcement Learning (PbRL) with human feedback. We consider the general reward setting where the reward can be defined over the whole trajectory. We introduce a new single-policy concentrability coefficient, which can be upper bounded by the per-trajectory concentrability.
  arXiv  Detail & Related papers  (2023-05-24T07:11:26Z)
• Truncating Trajectories in Monte Carlo Reinforcement Learning [48.97155920826079]
  In Reinforcement Learning (RL), an agent acts in an unknown environment to maximize the expected cumulative discounted sum of an external reward signal. We propose an a-priori budget allocation strategy that leads to the collection of trajectories of different lengths. We show that an appropriate truncation of the trajectories can succeed in improving performance.
  arXiv  Detail & Related papers  (2023-05-07T19:41:57Z)
• Principled Reinforcement Learning with Human Feedback from Pairwise or $K$-wise Comparisons [79.98542868281473]
  We provide a theoretical framework for Reinforcement Learning with Human Feedback (RLHF) We show that when training a policy based on the learned reward model, MLE fails while a pessimistic MLE provides policies with improved performance under certain coverage assumptions.
  arXiv  Detail & Related papers  (2023-01-26T18:07:21Z)
• False Correlation Reduction for Offline Reinforcement Learning [115.11954432080749]
  We propose falSe COrrelation REduction (SCORE) for offline RL, a practically effective and theoretically provable algorithm. We empirically show that SCORE achieves the SoTA performance with 3.1x acceleration on various tasks in a standard benchmark (D4RL)
  arXiv  Detail & Related papers  (2021-10-24T15:34:03Z)

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.