MetaRM: Shifted Distributions Alignment via Meta-Learning

• URL: http://arxiv.org/abs/2405.00438v1
• Date: Wed, 1 May 2024 10:43:55 GMT
• Title: MetaRM: Shifted Distributions Alignment via Meta-Learning
• Authors: Shihan Dou, Yan Liu, Enyu Zhou, Tianlong Li, Haoxiang Jia, Limao Xiong, Xin Zhao, Junjie Ye, Rui Zheng, Tao Gui, Qi Zhang, Xuanjing Huang,
• Abstract summary: Reinforcement Learning from Human Feedback (RLHF) in language model alignment is critically dependent on the capability of the reward model (RM) We introduce MetaRM, a method leveraging meta-learning to align the RM with the shifted environment distribution. Extensive experiments demonstrate that MetaRM significantly improves the RM's distinguishing ability in iterative RLHF optimization.
• Score: 52.94381279744458
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The success of Reinforcement Learning from Human Feedback (RLHF) in language model alignment is critically dependent on the capability of the reward model (RM). However, as the training process progresses, the output distribution of the policy model shifts, leading to the RM's reduced ability to distinguish between responses. This issue is further compounded when the RM, trained on a specific data distribution, struggles to generalize to examples outside of that distribution. These two issues can be united as a challenge posed by the shifted distribution of the environment. To surmount this challenge, we introduce MetaRM, a method leveraging meta-learning to align the RM with the shifted environment distribution. MetaRM is designed to train the RM by minimizing data loss, particularly for data that can improve the differentiation ability to examples of the shifted target distribution. Extensive experiments demonstrate that MetaRM significantly improves the RM's distinguishing ability in iterative RLHF optimization, and also provides the capacity to identify subtle differences in out-of-distribution samples.

Related papers

• Semi-Supervised Reward Modeling via Iterative Self-Training [52.48668920483908]
  We propose Semi-Supervised Reward Modeling (SSRM), an approach that enhances RM training using unlabeled data. We demonstrate that SSRM significantly improves reward models without incurring additional labeling costs. Overall, SSRM substantially reduces the dependency on large volumes of human-annotated data, thereby decreasing the overall cost and time involved in training effective reward models.
  arXiv  Detail & Related papers  (2024-09-10T22:57:58Z)
• A Distributional Analogue to the Successor Representation [54.99439648059807]
  This paper contributes a new approach for distributional reinforcement learning. It elucidates a clean separation of transition structure and reward in the learning process. As an illustration, we show that it enables zero-shot risk-sensitive policy evaluation.
  arXiv  Detail & Related papers  (2024-02-13T15:35:24Z)
• Distribution-Dependent Rates for Multi-Distribution Learning [26.38831409926518]
  Recent multi-distribution learning framework tackles this objective in a dynamic interaction with the environment. We provide distribution-dependent guarantees in the MDL regime, that scale with suboptimality gaps and result in superior dependence on the sample size. We devise an adaptive optimistic algorithm, LCB-DR, that showcases enhanced dependence on the gaps, mirroring the contrast between uniform and optimistic allocation in the multi-armed bandit literature.
  arXiv  Detail & Related papers  (2023-12-20T15:50:16Z)
• Noise Distribution Decomposition based Multi-Agent Distributional Reinforcement Learning [15.82785057592436]
  Multi-Agent Reinforcement Learning (MARL) is more susceptible to noise due to the interference among intelligent agents. We propose a novel decomposition-based multi-agent distributional RL method by approxing the globally shared noisy reward. We also verify the effectiveness of the proposed method through extensive simulation experiments with noisy rewards.
  arXiv  Detail & Related papers  (2023-12-12T07:24:15Z)
• MDDL: A Framework for Reinforcement Learning-based Position Allocation in Multi-Channel Feed [14.8342816935259]
  We propose a framework named Multi-Distribution Data Learning (MDDL) to address the challenge of effectively utilizing both strategy and random data for training RL models. MDDL incorporates a novel imitation learning signal to mitigate overestimation problems in strategy data and maximizes the RL signal for random data to facilitate effective learning. MDDL has been fully deployed on the Meituan food delivery platform and currently serves over 300 million users.
  arXiv  Detail & Related papers  (2023-04-17T07:25:58Z)
• Learning Optimal Features via Partial Invariance [18.552839725370383]
  Invariant Risk Minimization (IRM) is a popular framework that aims to learn robust models from multiple environments. We show that IRM can over-constrain the predictor and to remedy this, we propose a relaxation via $textitpartial invariance$. Several experiments, conducted both in linear settings as well as with deep neural networks on tasks over both language and image data, allow us to verify our conclusions.
  arXiv  Detail & Related papers  (2023-01-28T02:48:14Z)
• From Cloze to Comprehension: Retrofitting Pre-trained Masked Language Model to Pre-trained Machine Reader [130.45769668885487]
  Pre-trained Machine Reader (PMR) is a novel method for retrofitting masked language models (MLMs) to pre-trained machine reading comprehension (MRC) models without acquiring labeled data. To build the proposed PMR, we constructed a large volume of general-purpose and high-quality MRC-style training data. PMR has the potential to serve as a unified model for tackling various extraction and classification tasks in the MRC formulation.
  arXiv  Detail & Related papers  (2022-12-09T10:21:56Z)
• On Reinforcement Learning and Distribution Matching for Fine-Tuning Language Models with no Catastrophic Forgetting [5.5302127686575435]
  Two main paradigms have emerged to tackle this challenge: Reward Maximization (RM) and, more recently, Distribution Matching (DM) We show that methods such as KL-control developed for RM can also be construed as belonging to DM. We leverage connections between the two paradigms to import the concept of baseline into DM methods.
  arXiv  Detail & Related papers  (2022-06-01T20:54:41Z)
• Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalization [89.73665256847858]
  We show that out-of-distribution performance is strongly correlated with in-distribution performance for a wide range of models and distribution shifts. Specifically, we demonstrate strong correlations between in-distribution and out-of-distribution performance on variants of CIFAR-10 & ImageNet. We also investigate cases where the correlation is weaker, for instance some synthetic distribution shifts from CIFAR-10-C and the tissue classification dataset Camelyon17-WILDS.
  arXiv  Detail & Related papers  (2021-07-09T19:48:23Z)
• When Relation Networks meet GANs: Relation GANs with Triplet Loss [110.7572918636599]
  Training stability is still a lingering concern of generative adversarial networks (GANs) In this paper, we explore a relation network architecture for the discriminator and design a triplet loss which performs better generalization and stability. Experiments on benchmark datasets show that the proposed relation discriminator and new loss can provide significant improvement on variable vision tasks.
  arXiv  Detail & Related papers  (2020-02-24T11:35:28Z)

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.