Proximal Policy Optimization for Improved Convergence in IRGAN

TL;DR

This paper introduces an improved IRGAN training method using proximal policy optimization and Gumbel-Softmax sampling, leading to more stable convergence and better performance across IR tasks.

Contribution

It proposes a novel training objective and algorithm for IRGAN that enhances stability and convergence, with empirical validation on benchmark datasets.

Findings

Enhanced convergence stability over original IRGAN

Superior performance on multiple IR benchmark datasets

Effective use of proximal policy optimization in IR training

Abstract

IRGAN is an information retrieval (IR) modeling approach that uses a theoretical minimax game between a generative and a discriminative model to iteratively optimize both of them, hence unifying the generative and discriminative approaches. Despite significant performance improvements in several information retrieval tasks, IRGAN training is an unstable process, and the solution varies largely with the random parameter initialization. In this work, we present an improved training objective based on proximal policy optimization objective and Gumbel-Softmax based sampling for the generator. We also propose a modified training algorithm which takes a single gradient update on both the generator as well as discriminator for each iteration step. We present empirical evidence of the improved convergence of the proposed model over the original IRGAN and a comparison on three different IR tasks on benchmark datasets is also discussed, emphasizing the proposed model's superior performance.