Timezone: »

 
Poster
Latent Attention For If-Then Program Synthesis
Chang Liu · Xinyun Chen · Richard Shin · Mingcheng Chen · Dawn Song

Wed Dec 07 09:00 AM -- 12:30 PM (PST) @ Area 5+6+7+8 #50
in Wednesday Posters »

Automatic translation from natural language descriptions into programs is a long-standing challenging problem. In this work, we consider a simple yet important sub-problem: translation from textual descriptions to If-Then programs. We devise a novel neural network architecture for this task which we train end-to-end. Specifically, we introduce Latent Attention, which computes multiplicative weights for the words in the description in a two-stage process with the goal of better leveraging the natural language structures that indicate the relevant parts for predicting program elements. Our architecture reduces the error rate by 28.57% compared to prior art. We also propose a one-shot learning scenario of If-Then program synthesis and simulate it with our existing dataset. We demonstrate a variation on the training procedure for this scenario that outperforms the original procedure, significantly closing the gap to the model trained with all data.

Keywords: Deep Learning or Neural Networks · Multi-task and Transfer Learning · (Application) Natural Language and Text Processing · (Cognitive/Neuroscience) Language
[ PDF] 

Author Information

Chang Liu (University of Maryland)
Xinyun Chen (Shanghai Jiaotong University)
Richard Shin (UC Berkeley)
Mingcheng Chen (University of Illinois)
Dawn Song (UC Berkeley)

More from the Same Authors