Trainable Greedy Decoding for Neural Machine Translation

Abstract

Recent research in neural machine translation has largely focused on two aspects; neural network architectures and end-to-end learning algorithms. The problem of decoding, however, has received relatively little attention from the research community. In this paper, we solely focus on the problem of decoding given a trained neural machine translation model. Instead of trying to build a new decoding algorithm for any specific decoding objective, we propose the idea of trainable decoding algorithm in which we train a decoding algorithm to find a translation that maximizes an arbitrary decoding objective. More specifically, we design an actor that observes and manipulates the hidden state of the neural machine translation decoder and propose to train it using a variant of deterministic policy gradient. We extensively evaluate the proposed algorithm using four language pairs and two decoding objectives and show that we can indeed train a trainable greedy decoder that generates a better translation (in terms of a target decoding objective) with minimal computational overhead.

10 pages

References in corpus (19)

• Continuous control with deep reinforcement learning
• Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
• ADADELTA: An Adaptive Learning Rate Method
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Sequence Level Training with Recurrent Neural Networks
• Learning Continuous Control Policies by Stochastic Value Gradients
• Neural Machine Translation in Linear Time
• Toward Multilingual Neural Machine Translation with Universal Encoder and Decoder
• A Simple, Fast Diverse Decoding Algorithm for Neural Generation
• An Actor-Critic Algorithm for Sequence Prediction
• Character-based Neural Machine Translation
• Modeling Coverage for Neural Machine Translation
• Sequence-to-Sequence Learning as Beam-Search Optimization
• Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation
• Neural Machine Translation with Reconstruction
• A Convolutional Encoder Model for Neural Machine Translation
• Learning to Decode for Future Success
• Noisy Parallel Approximate Decoding for Conditional Recurrent Language Model
• Minimum Risk Training for Neural Machine Translation

Cited by in corpus (6)

• Plug and Play Language Models: A Simple Approach to Controlled Text Generation
• Named Entity Recognition with stack residual LSTM and trainable bias decoding
• Controlling Decoding for More Abstractive Summaries with Copy-Based Networks
• Non-Autoregressive Neural Dialogue Generation
• Later-stage Minimum Bayes-Risk Decoding for Neural Machine Translation
• Reverse Engineering Configurations of Neural Text Generation Models