papers AI Learner
The Github is limit! Click to go to the new site.

Optimal Completion Distillation for Sequence Learning

2019-01-14
Sara Sabour, William Chan, Mohammad Norouzi

Abstract

We present Optimal Completion Distillation (OCD), a training procedure for optimizing sequence to sequence models based on edit distance. OCD is efficient, has no hyper-parameters of its own, and does not require pretraining or joint optimization with conditional log-likelihood. Given a partial sequence generated by the model, we first identify the set of optimal suffixes that minimize the total edit distance, using an efficient dynamic programming algorithm. Then, for each position of the generated sequence, we use a target distribution that puts equal probability on the first token of all the optimal suffixes. OCD achieves the state-of-the-art performance on end-to-end speech recognition, on both Wall Street Journal and Librispeech datasets, achieving $9.3\%$ WER and $4.5\%$ WER respectively.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1810.01398

PDF

http://arxiv.org/pdf/1810.01398


Similar Posts

Comments