Accelerated Reinforcement Learning for Sentence Generation by Vocabulary Prediction

Abstract

A major obstacle in reinforcement learning-based sentence generation is the large action space whose size is equal to the vocabulary size of the target-side language. To improve the efficiency of reinforcement learning, we present a novel approach for reducing the action space based on dynamic vocabulary prediction. Our method first predicts a fixed-size small vocabulary for each input to generate its target sentence. The input-specific vocabularies are then used at supervised and reinforcement learning steps, and also at test time. In our experiments on six machine translation and two image captioning datasets, our method achieves faster reinforcement learning ($\sim$2.7x faster) with less GPU memory ($\sim$2.3x less) than the full-vocabulary counterpart. The reinforcement learning with our method consistently leads to significant improvement of BLEU scores, and the scores are equal to or better than those of baselines using the full vocabularies, with faster decoding time ($\sim$3x faster) on CPUs.

NAACL2019 camera ready (mini-batch splitting is added)

References in corpus (9)

• Sequence to Sequence Learning with Neural Networks
• Improving neural networks by preventing co-adaptation of feature detectors
• Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
• Recurrent Neural Network Regularization
• A Deep Reinforced Model for Abstractive Summarization
• Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modeling
• Analyzing Uncertainty in Neural Machine Translation
• Neural Machine Translation with Word Predictions
• Neural Machine Translation via Binary Code Prediction