Neural Sketch Learning for Conditional Program Generation

Abstract

We study the problem of generating source code in a strongly typed, Java-like programming language, given a label (for example a set of API calls or types) carrying a small amount of information about the code that is desired. The generated programs are expected to respect a "realistic" relationship between programs and labels, as exemplified by a corpus of labeled programs available during training. Two challenges in such conditional program generation are that the generated programs must satisfy a rich set of syntactic and semantic constraints, and that source code contains many low-level features that impede learning. We address these problems by training a neural generator not on code but on program sketches, or models of program syntax that abstract out names and operations that do not generalize across programs. During generation, we infer a posterior distribution over sketches, then concretize samples from this distribution into type-safe programs using combinatorial techniques. We implement our ideas in a system for generating API-heavy Java code, and show that it can often predict the entire body of a method given just a few API calls or data types that appear in the method.

Cited by in corpus (17)

• Review: Deep Learning in Electron Microscopy
• Programmatically Interpretable Reinforcement Learning
• Structuring the Synthesis of Heap-Manipulating Programs - Extended Version
• SketchGraphs: A Large-Scale Dataset for Modeling Relational Geometry in Computer-Aided Design
• Learning Compositional Rules via Neural Program Synthesis
• Program Synthesis with Large Language Models
• Neural Attribute Machines for Program Generation
• ENCORE: Ensemble Learning using Convolution Neural Machine Translation for Automatic Program Repair
• Learning the Relation between Code Features and Code Transforms with Structured Prediction
• Imitation-Projected Programmatic Reinforcement Learning
• Learning Differentiable Programs with Admissible Neural Heuristics
• Latent Programmer: Discrete Latent Codes for Program Synthesis
• Discrete Object Generation with Reversible Inductive Construction
• Robot Action Selection Learning via Layered Dimension Informed Program Synthesis
• PSCS: A Path-based Neural Model for Semantic Code Search
• Targeted Synthesis for Programming with Data Invariants
• Neural Program Generation Modulo Static Analysis