GPG: Generalized Policy Gradient Theorem for Transformer-based Policies

Hangyu Mao; Guangting Dong; Zhicheng Dou

Paper

GPG: Generalized Policy Gradient Theorem for Transformer-based Policies

Abstract

We present the Generalized Policy Gradient (GPG) Theorem, specifically designed for Transformer-based policies. Notably, we demonstrate that both standard Policy Gradient Theorem and GRPO emerge as special cases within our GPG framework. Furthermore, we explore its practical applications in training Large Language Models (LLMs), offering new insights into efficient policy optimization.