Strategyproof Reinforcement Learning from Human Feedback

• URL: http://arxiv.org/abs/2503.09561v1
• Date: Wed, 12 Mar 2025 17:25:52 GMT
• Title: Strategyproof Reinforcement Learning from Human Feedback
• Authors: Thomas Kleine Buening, Jiarui Gan, Debmalya Mandal, Marta Kwiatkowska,
• Abstract summary: We show that existing RLHF methods are not strategyproof.<n>We also find that any strategyproof RLHF algorithm must perform $k$-times worse than the optimal policy.
• Score: 27.129105195239465
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We study Reinforcement Learning from Human Feedback (RLHF), where multiple individuals with diverse preferences provide feedback strategically to sway the final policy in their favor. We show that existing RLHF methods are not strategyproof, which can result in learning a substantially misaligned policy even when only one out of $k$ individuals reports their preferences strategically. In turn, we also find that any strategyproof RLHF algorithm must perform $k$-times worse than the optimal policy, highlighting an inherent trade-off between incentive alignment and policy alignment. We then propose a pessimistic median algorithm that, under appropriate coverage assumptions, is approximately strategyproof and converges to the optimal policy as the number of individuals and samples increases.

Related papers

• Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning [55.65738319966385]
  We propose a novel online algorithm, iterative Nash policy optimization (INPO)<n>Unlike previous methods, INPO bypasses the need for estimating the expected win rate for individual responses.<n>With an LLaMA-3-8B-based SFT model, INPO achieves a 42.6% length-controlled win rate on AlpacaEval 2.0 and a 37.8% win rate on Arena-Hard.
  arXiv  Detail & Related papers  (2024-06-30T08:00:34Z)
• POTEC: Off-Policy Learning for Large Action Spaces via Two-Stage Policy Decomposition [40.851324484481275]
  We study off-policy learning of contextual bandit policies in large discrete action spaces. We propose a novel two-stage algorithm, called Policy Optimization via Two-Stage Policy Decomposition. We show that POTEC provides substantial improvements in OPL effectiveness particularly in large and structured action spaces.
  arXiv  Detail & Related papers  (2024-02-09T03:01:13Z)
• 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)
• Importance Weighted Actor-Critic for Optimal Conservative Offline Reinforcement Learning [23.222448307481073]
  We propose a new practical algorithm for offline reinforcement learning (RL) in complex environments with insufficient data coverage. Our algorithm combines the marginalized importance sampling framework with the actor-critic paradigm. We provide both theoretical analysis and experimental results to validate the effectiveness of our proposed algorithm.
  arXiv  Detail & Related papers  (2023-01-30T07:53:53Z)
• Policy learning "without" overlap: Pessimism and generalized empirical Bernstein's inequality [94.89246810243053]
  This paper studies offline policy learning, which aims at utilizing observations collected a priori to learn an optimal individualized decision rule. Existing policy learning methods rely on a uniform overlap assumption, i.e., the propensities of exploring all actions for all individual characteristics must be lower bounded. We propose Pessimistic Policy Learning (PPL), a new algorithm that optimize lower confidence bounds (LCBs) instead of point estimates.
  arXiv  Detail & Related papers  (2022-12-19T22:43:08Z)
• Coordinate Ascent for Off-Policy RL with Global Convergence Guarantees [8.610425739792284]
  We revisit the domain of off-policy policy optimization in RL. One commonly-used approach is to leverage the off-policy policy gradient to optimize a surrogate objective. This approach has been shown to suffer from the distribution mismatch issue.
  arXiv  Detail & Related papers  (2022-12-10T07:47:04Z)
• Strategic Decision-Making in the Presence of Information Asymmetry: Provably Efficient RL with Algorithmic Instruments [55.41685740015095]
  We study offline reinforcement learning under a novel model called strategic MDP. We propose a novel algorithm, Pessimistic policy Learning with Algorithmic iNstruments (PLAN)
  arXiv  Detail & Related papers  (2022-08-23T15:32:44Z)
• CAMEO: Curiosity Augmented Metropolis for Exploratory Optimal Policies [62.39667564455059]
  We consider and study a distribution of optimal policies. In experimental simulations we show that CAMEO indeed obtains policies that all solve classic control problems. We further show that the different policies we sample present different risk profiles, corresponding to interesting practical applications in interpretability.
  arXiv  Detail & Related papers  (2022-05-19T09:48:56Z)
• Attacking and Defending Deep Reinforcement Learning Policies [3.6985039575807246]
  We study robustness of DRL policies to adversarial attacks from the perspective of robust optimization. We propose a greedy attack algorithm, which tries to minimize the expected return of the policy without interacting with the environment, and a defense algorithm, which performs adversarial training in a max-min form.
  arXiv  Detail & Related papers  (2022-05-16T12:47:54Z)
• Understanding the Effect of Stochasticity in Policy Optimization [86.7574122154668]
  We show that the preferability of optimization methods depends critically on whether exact gradients are used. Second, to explain these findings we introduce the concept of committal rate for policy optimization. Third, we show that in the absence of external oracle information, there is an inherent trade-off between exploiting geometry to accelerate convergence versus achieving optimality almost surely.
  arXiv  Detail & Related papers  (2021-10-29T06:35:44Z)

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.