DeepSynth: Scaling Neural Program Synthesis with Distribution-based Search

Writing software is tedious, error-prone, and accessible only to a small share of the population -- yet coding grows increasingly important as the digital world plays larger and larger roles in peoples’ lives.
Programming by example seeks to make programming more reliable and accessible by allowing non-technical users to specify programs only from pairs of input-output examples.
`DeepSynth` is a general purpose programming by example tool, which combines predictions from neural networks with symbolic methods to generate programs.