DeepAI AI Chat
Log In Sign Up

Reinforcement Learning based Curriculum Optimization for Neural Machine Translation

by   Gaurav Kumar, et al.

We consider the problem of making efficient use of heterogeneous training data in neural machine translation (NMT). Specifically, given a training dataset with a sentence-level feature such as noise, we seek an optimal curriculum, or order for presenting examples to the system during training. Our curriculum framework allows examples to appear an arbitrary number of times, and thus generalizes data weighting, filtering, and fine-tuning schemes. Rather than relying on prior knowledge to design a curriculum, we use reinforcement learning to learn one automatically, jointly with the NMT system, in the course of a single training run. We show that this approach can beat uniform and filtering baselines on Paracrawl and WMT English-to-French datasets by up to +3.4 BLEU, and match the performance of a hand-designed, state-of-the-art curriculum.


Curriculum Learning and Minibatch Bucketing in Neural Machine Translation

We examine the effects of particular orderings of sentence pairs on the ...

Data Selection Curriculum for Neural Machine Translation

Neural Machine Translation (NMT) models are typically trained on heterog...

Competence-based Curriculum Learning for Neural Machine Translation

Current state-of-the-art NMT systems use large neural networks that are ...

Self-Guided Curriculum Learning for Neural Machine Translation

In the field of machine learning, the well-trained model is assumed to b...

Learning a Multitask Curriculum for Neural Machine Translation

Existing curriculum learning research in neural machine translation (NMT...

TSO: Curriculum Generation using continuous optimization

The training of deep learning models poses vast challenges of including ...

Curriculum Q-Learning for Visual Vocabulary Acquisition

The structure of curriculum plays a vital role in our learning process, ...