A bi-objective perspective on controllable language models: reward dropout improves off-policy control performance
Primary Area: general machine learning (i.e., none of the above)
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Keywords: Controllable Language Model, Language Model, Reinforcement Learning, Off-policy RL, Bi-objective Problem, Pareto Optimization
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
Abstract: We study the theoretical aspects of CLMs (Controllable Language Models) from a bi-objective optimization perspective. Specifically, we consider the CLMs as an off-policy RL problem that requires simultaneously maximizing the reward and likelihood objectives. Our main contribution consists of three parts. First, we establish the theoretical foundations of CLM by presenting reward upper bound and Pareto improvement/optimality conditions. Second, we analyze conditions that improve and violate Pareto optimality itself, respectively. Finally, we propose Reward Dropout, a simple yet powerful method to guarantee policy improvement based on a Pareto improvement condition. Our theoretical outcomes are supported by not only deductive proofs but also empirical results. The performance of Reward Dropout was evaluated on five CLM benchmark datasets, and it turns out that the Reward Dropout significantly improves the performance of CLMs.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 7118
Loading