Reinforcement Learning based Curriculum Optimization for Neural Machine Translation
This addresses the challenge of efficiently using heterogeneous data in NMT, offering an automated alternative to manual curriculum design, though it is incremental as it builds on existing curriculum learning methods.
The paper tackles the problem of optimizing training data order for neural machine translation using reinforcement learning to automatically learn a curriculum, achieving up to +3.4 BLEU improvement over baselines and matching hand-designed state-of-the-art performance.
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.