Timezone: »

 
Poster
Navigating with Graph Representations for Fast and Scalable Decoding of Neural Language Models
Minjia Zhang · Wenhan Wang · Xiaodong Liu · Jianfeng Gao · Yuxiong He

Tue Dec 04 07:45 AM -- 09:45 AM (PST) @ Room 210 #73
in Tue Poster Session A »

Neural language models (NLMs) have recently gained a renewed interest by achieving state-of-the-art performance across many natural language processing (NLP) tasks. However, NLMs are very computationally demanding largely due to the computational cost of the decoding process, which consists of a softmax layer over a large vocabulary.We observe that in the decoding of many NLP tasks, only the probabilities of the top-K hypotheses need to be calculated preciously and K is often much smaller than the vocabulary size. This paper proposes a novel softmax layer approximation algorithm, called Fast Graph Decoder (FGD), which quickly identifies, for a given context, a set of K words that are most likely to occur according to a NLM. We demonstrate that FGD reduces the decoding time by an order of magnitude while attaining close to the full softmax baseline accuracy on neural machine translation and language modeling tasks. We also prove the theoretical guarantee on the softmax approximation quality.

Keywords: Deep Learning · Natural Language Processing · Efficient Inference Methods
[ Paper]  [ Poster] 

Author Information

Minjia Zhang (Microsoft)
Wenhan Wang (Microsoft)
Xiaodong Liu (Microsoft)
Jianfeng Gao (Microsoft Research, Redmond, WA)
Yuxiong He (Microsoft)

More from the Same Authors