Image Captioning based on Deep Reinforcement Learning

TL;DR

This paper introduces a novel deep reinforcement learning architecture for image captioning, utilizing policy and value networks to improve caption quality on the Microsoft COCO dataset.

Contribution

The paper proposes a new deep reinforcement learning approach with policy and value networks for image captioning, differing from traditional sequential models like RNNs.

Findings

Effective caption generation verified on Microsoft COCO dataset

Deep reinforcement learning improves caption quality

Collaborative policy and value networks enhance performance

Abstract

Recently it has shown that the policy-gradient methods for reinforcement learning have been utilized to train deep end-to-end systems on natural language processing tasks. What's more, with the complexity of understanding image content and diverse ways of describing image content in natural language, image captioning has been a challenging problem to deal with. To the best of our knowledge, most state-of-the-art methods follow a pattern of sequential model, such as recurrent neural networks (RNN). However, in this paper, we propose a novel architecture for image captioning with deep reinforcement learning to optimize image captioning tasks. We utilize two networks called "policy network" and "value network" to collaboratively generate the captions of images. The experiments are conducted on Microsoft COCO dataset, and the experimental results have verified the effectiveness of the proposed method.