Inductive program synthesis over noisy data

Shivam Handa,Martin C Rinard

doi:10.1145/3368089.3409732

Abstract

We present a new framework and associated synthesis algorithms for program synthesis over noisy data, i.e., data that may contain incorrect/corrupted input-output examples. This framework is based on an extension of finite tree automata called {\em weighted finite tree automata}. We show how to apply this framework to formulate and solve a variety of program synthesis problems over noisy data. Results from our implemented system running on problems from the SyGuS 2018 benchmark suite highlight its ability to successfully synthesize programs in the face of noisy data sets, including the ability to synthesize a correct program even when every input-output example in the data set is corrupted.

Highlights

In recent years there has been significant interest in learning programs from input-output examples
The results indicate that the technique is effective at solving program synthesis problems over strings with modestly sized solutions even in the presence of substantial noise
For discrete noise sources and a loss function that is a good match for the noise source, the technique is typically able to extract enough information left intact in corrupted outputs to synthesize a correct program even when all outputs are corrupted

Summary

Introduction

In recent years there has been significant interest in learning programs from input-output examples These techniques have been successfully used to synthesize programs for domains such as string and format transformations [10, 18], data wrangling [7], data completion [20], and data structure manipulation [8, 12, 21]. Even though these efforts have been largely successful, they do not aspire to work with noisy data sets that may contain corrupted input-output examples.

Results

Discussion

Conclusion