# Is RLHF More Difficult than Standard RL?

arXiv: https://arxiv.org/abs/2306.14111 · HTML: https://mlsanity.com/paper/2306.14111/

## Key facts

- Cited by 96 papers in the map (data through 14 Oct 2026).
- Placed in Reward modeling and RLHF, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 32 of its 56 references that are in the map.
- First posted on arXiv 25 Jun 2023.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 30 of 32)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [Dota 2 with Large Scale Deep Reinforcement Learning](https://mlsanity.com/paper/1912.06680/) (2019)
- [Provably Efficient Reinforcement Learning with Linear Function Approximation](https://mlsanity.com/paper/1907.05388/) (2019)
- [Minimax Regret Bounds for Reinforcement Learning](https://mlsanity.com/paper/1703.05449/) (2017)
- [Aligning Text-to-Image Models using Human Feedback](https://mlsanity.com/paper/2302.12192/) (2023)
- [Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization](https://mlsanity.com/paper/1603.00448/) (2016)
- [Reward learning from human preferences and demonstrations in Atari](https://mlsanity.com/paper/1811.06521/) (2018)
- [Principled Reinforcement Learning with Human Feedback from Pairwise or K-wise Comparisons](https://mlsanity.com/paper/2301.11270/) (2023)
- [Bellman Eluder Dimension: New Rich Classes of RL Problems, and Sample-Efficient Algorithms](https://mlsanity.com/paper/2102.00815/) (2021)
- [The Statistical Complexity of Interactive Decision Making](https://mlsanity.com/paper/2112.13487/) (2021)
- [Learning Near Optimal Policies with Low Inherent Bellman Error](https://mlsanity.com/paper/2003.00153/) (2020)
- [Reward-Free Exploration for Reinforcement Learning](https://mlsanity.com/paper/2002.02794/) (2020)
- [Deep TAMER: Interactive Agent Shaping in High-Dimensional State Spaces](https://mlsanity.com/paper/1709.10163/) (2017)
- [Dueling RL: Reinforcement Learning with Trajectory Preferences](https://mlsanity.com/paper/2111.04850/) (2021)
- [Preference-based Online Learning with Dueling Bandits: A Survey](https://mlsanity.com/paper/1807.11398/) (2018)
- [Contextual Dueling Bandits](https://mlsanity.com/paper/1502.06362/) (2015)
- [Human-in-the-loop: Provably Efficient Preference-based Reinforcement Learning with General Function Approximation](https://mlsanity.com/paper/2205.11140/) (2022)
- [Preference-based Reinforcement Learning with Finite-Time Guarantees](https://mlsanity.com/paper/2006.08910/) (2020)
- [Reducing Dueling Bandits to Cardinal Bandits](https://mlsanity.com/paper/1405.3396/) (2014)
- [Dueling Posterior Sampling for Preference-Based Reinforcement Learning](https://mlsanity.com/paper/1908.01289/) (2019)
- [DQN-TAMER: Human-in-the-Loop Reinforcement Learning with Intractable Feedback](https://mlsanity.com/paper/1810.11748/) (2018)
- [Learning Trajectory Preferences for Manipulators via Iterative Improvement](https://mlsanity.com/paper/1306.6294/) (2013)
- [PAC Battling Bandits in the Plackett-Luce Model](https://mlsanity.com/paper/1808.04008/) (2018)
- [Sample-Efficient Reinforcement Learning of Partially Observable Markov Games](https://mlsanity.com/paper/2206.01315/) (2022)
- [Improving Multimodal Interactive Agents with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2211.11602/) (2022)
- [Improved Regret for Efficient Online Reinforcement Learning with Linear Function Approximation](https://mlsanity.com/paper/2301.13087/) (2023)

## Built on by (most cited first, 30 of 96)

- [Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-Constraint](https://mlsanity.com/paper/2312.11456/) (2023)
- [Self-Play Preference Optimization for Language Model Alignment](https://mlsanity.com/paper/2405.00675/) (2024)
- [Nash Learning from Human Feedback](https://mlsanity.com/paper/2312.00886/) (2023)
- [Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferences](https://mlsanity.com/paper/2404.03715/) (2024)
- [Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning](https://mlsanity.com/paper/2408.10075/) (2024)
- [Arithmetic Control of LLMs for Diverse User Preferences: Directional Preference Alignment with Multi-Objective Rewards](https://mlsanity.com/paper/2402.18571/) (2024)
- [DPO Meets PPO: Reinforced Token Optimization for RLHF](https://mlsanity.com/paper/2404.18922/) (2024)
- [Provably Mitigating Overoptimization in RLHF: Your SFT Loss is Implicitly an Adversarial Regularizer](https://mlsanity.com/paper/2405.16436/) (2024)
- [Exploratory Preference Optimization: Harnessing Implicit Q*-Approximation for Sample-Efficient RLHF](https://mlsanity.com/paper/2405.21046/) (2024)
- [RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold](https://mlsanity.com/paper/2406.14532/) (2024)
- [Is Best-of-N the Best of Them? Coverage, Scaling, and Optimality in Inference-Time Alignment](https://mlsanity.com/paper/2503.21878/) (2025)
- [Correcting the Mythos of KL-Regularization: Direct Alignment without Overoptimization via Chi-Squared Preference Optimization](https://mlsanity.com/paper/2407.13399/) (2024)
- [REBEL: Reinforcement Learning via Regressing Relative Rewards](https://mlsanity.com/paper/2404.16767/) (2024)
- [Multi-turn Reinforcement Learning from Preference Human Feedback](https://mlsanity.com/paper/2405.14655/) (2024)
- [Building Math Agents with Multi-Turn Iterative Preference Learning](https://mlsanity.com/paper/2409.02392/) (2024)
- [Self-Exploring Language Models: Active Preference Elicitation for Online Alignment](https://mlsanity.com/paper/2405.19332/) (2024)
- [Online Iterative Reinforcement Learning from Human Feedback with General Preference Model](https://mlsanity.com/paper/2402.07314/) (2024)
- [Making RL with Preference-based Feedback Efficient via Randomization](https://mlsanity.com/paper/2310.14554/) (2023)
- [Reinforcement Learning from Human Feedback with Active Queries](https://mlsanity.com/paper/2402.09401/) (2024)
- [A Survey of Recent Backdoor Attacks and Defenses in Large Language Models](https://mlsanity.com/paper/2406.06852/) (2024)
- [Sharp Analysis for KL-Regularized Contextual Bandits and RLHF](https://mlsanity.com/paper/2411.04625/) (2024)
- [Regressing the Relative Future: Efficient Policy Optimization for Multi-turn RLHF](https://mlsanity.com/paper/2410.04612/) (2024)
- [Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment](https://mlsanity.com/paper/2410.02197/) (2024)
- [Optimal Design for Reward Modeling in RLHF](https://mlsanity.com/paper/2410.17055/) (2024)
- [PerfCodeGen: Improving Performance of LLM Generated Code with Execution Feedback](https://mlsanity.com/paper/2412.03578/) (2024)
- [Logarithmic Regret for Online KL-Regularized Reinforcement Learning](https://mlsanity.com/paper/2502.07460/) (2025)
- [Distortion of AI Alignment: Does Preference Optimization Optimize for Preferences?](https://mlsanity.com/paper/2505.23749/) (2025)
- [Corruption Robust Offline Reinforcement Learning with Human Feedback](https://mlsanity.com/paper/2402.06734/) (2024)
- [Optimal Design for Human Preference Elicitation](https://mlsanity.com/paper/2404.13895/) (2024)
- [Principled Preferential Bayesian Optimization](https://mlsanity.com/paper/2402.05367/) (2024)

## Abstract (arXiv)

Reinforcement learning from Human Feedback (RLHF) learns from preference signals, while standard Reinforcement Learning (RL) directly learns from reward signals. Preferences arguably contain less information than rewards, which makes preference-based RL seemingly more difficult. This paper theoretically proves that, for a wide range of preference models, we can solve preference-based RL directly using existing algorithms and techniques for reward-based RL, with small or no extra costs. Specifically, (1) for preferences that are drawn from reward-based probabilistic models, we reduce the problem to robust reward-based RL that can tolerate small errors in rewards; (2) for general arbitrary preferences where the objective is to find the von Neumann winner, we reduce the problem to multiagent reward-based RL which finds Nash equilibria for factored Markov games with a restricted set of policies. The latter case can be further reduced to adversarial MDP when preferences only depend on the final state. We instantiate all reward-based RL subroutines by concrete provable algorithms, and apply our theory to a large class of models including tabular MDPs and MDPs with generic function approximation. We further provide guarantees when K-wise comparisons are available.