# Program Synthesis Through Reinforcement Learning Guided Tree Search

arXiv: https://arxiv.org/abs/1806.02932 · HTML: https://mlsanity.com/paper/1806.02932/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Jun 2018.

## Built on by (most cited first, 7 of 7)

- [Planning with Large Language Models for Code Generation](https://mlsanity.com/paper/2303.05510/) (2023)
- [Learning Math Reasoning from Self-Sampled Correct and Partially-Correct Solutions](https://mlsanity.com/paper/2205.14318/) (2022)
- [B-Coder: Value-Based Deep Reinforcement Learning for Program Synthesis](https://mlsanity.com/paper/2310.03173/) (2023)
- [A Survey on Pretrained Language Models for Neural Code Intelligence](https://mlsanity.com/paper/2212.10079/) (2022)
- [From Reasoning to Code: GRPO Optimization for Underrepresented Languages](https://mlsanity.com/paper/2506.11027/) (2025)
- [Expected Return Causes Outcome-Level Mode Collapse in Reinforcement Learning and How to Fix It with Inverse Probability Scaling](https://mlsanity.com/paper/2601.21669/) (2026)
- [Program Generation from Diverse Video Demonstrations](https://mlsanity.com/paper/2302.00178/) (2023)

## Abstract (arXiv)

Program Synthesis is the task of generating a program from a provided specification. Traditionally, this has been treated as a search problem by the programming languages (PL) community and more recently as a supervised learning problem by the machine learning community. Here, we propose a third approach, representing the task of synthesizing a given program as a Markov decision process solvable via reinforcement learning(RL). From observations about the states of partial programs, we attempt to find a program that is optimal over a provided reward metric on pairs of programs and states. We instantiate this approach on a subset of the RISC-V assembly language operating on floating point numbers, and as an optimization inspired by search-based techniques from the PL community, we combine RL with a priority search tree. We evaluate this instantiation and demonstrate the effectiveness of our combined method compared to a variety of baselines, including a pure RL ablation and a state of the art Markov chain Monte Carlo search method on this task.