paper-with-me

Papers

Imitating Language via Scalable Inverse Reinforcement Learning

2024-09-02 · Markus Wulfmeier, Michael Bloesch, Nino Vieillard, Arun Ahuja, Jorg Bornschein, Sandy Huang, Artem Sokolov, Matt Barnes, Guillaume Desjardins, Alex Bewley, Sarah Maria Elisabeth Bechtle, Jost Tobias Springenberg, Nikola Momchev, Olivier Bachem, Matthieu Geist, Martin Riedmiller

The majority of language model training builds on imitation learning. It covers pretraining, supervised fine-tuning, and affects the starting conditions for reinforcement learning from human feedback (RLHF). The simplicity and scalability of maximum likelihood estimation (MLE) for next token prediction led to its role as predominant paradigm. However, the broader field of imitation learning can more effectively utilize the sequential structure underlying autoregressive generation. We focus on investigating the inverse reinforcement learning (IRL) perspective to imitation, extracting rewards and directly optimizing sequences instead of individual token likelihoods and evaluate its benefits for fine-tuning large language models. We provide a new angle, reformulating inverse soft-Q-learning as a temporal difference regularized extension of MLE. This creates a principled connection between MLE and IRL and allows trading off added complexity with increased performance and diversity of generations in the supervised fine-tuning (SFT) setting. We find clear advantages for IRL-based imitation, in particular for retaining diversity while maximizing task performance, rendering IRL a strong alternative on fixed SFT datasets even without online data generation. Our analysis of IRL-extracted reward functions further indicates benefits for more robust reward functions via tighter integration of supervised and preference-based LLM post-training.

📄 PDF Abstract BibTeX arXiv:2409.01369

Code (0)

등록된 구현이 없습니다.

Tasks

DiversityImitation LearningLanguage ModellingQ-Learningreinforcement-learningReinforcement Learning

Methods 이 논문이 사용한 방법론

SFT Shrink and Fine-Tune, or SFT, is a type of distillation that avoids explicit distillation by copying parameters to a student student model and then fine-tuning.…
Focus 설명 없음

Similar Papers 제목 키워드 기반

Generative Adversarial Imitation Learning

2016-06-10 · NeurIPS 2016 12 · Jonathan Ho, Stefano Ermon

Consider learning a policy from example expert behavior, without interaction with the expert or access to reinforcement signal. One approach is to recover the expert's cost function with inverse reinforcement learning, t…

Imitation Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Inverse Reinforcement Learning for Text Summarization

2022-12-19 · Yu Fu, Deyi Xiong, Yue Dong

We introduce inverse reinforcement learning (IRL) as an effective paradigm for training abstractive summarization models, imitating human summarization behaviors. Our IRL model estimates the reward function using a suite…

Abstractive Text Summarizationreinforcement-learningReinforcement LearningReinforcement Learning (RL)+1

Lifelong Inverse Reinforcement Learning

2022-07-01 · NeurIPS 2018 12 · Jorge A. Mendez, Shashank Shivkumar, Eric Eaton

Methods for learning from demonstration (LfD) have shown success in acquiring behavior policies by imitating a user. However, even for a single task, LfD may require numerous demonstrations. For versatile agents that mus…

Lifelong learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Learning Reasoning Rewards from Expert Demonstrations with Inverse Reinforcement Learning

2025-10-02 · Claudio Fanconi, Nicolás Astorga, Mihaela van der Schaar arxiv

Teaching large language models (LLMs) to reason during post-training typically relies on reinforcement learning with explicit outcome- or process-based reward functions. However, in many real-world settings, obtaining or…

Reinforcement Learning

Adversarial Exploration Strategy for Self-Supervised Imitation Learning

2019-05-01 · ICLR 2019 5 · Zhang-Wei Hong, Tsu-Jui Fu, Tzu-Yun Shann, Yi-Hsiang Chang 외

We present an adversarial exploration strategy, a simple yet effective imitation learning scheme that incentivizes exploration of an environment without any extrinsic reward or human demonstration. Our framework consists…

Deep Reinforcement LearningImitation LearningOpenAI GymReinforcement Learning