Reinforcement learning for mutation operator selection in automated program repair

Carol Hanna, Aymeric Blot, Justyna Petke

Automated Software Engineering 2025 article cited by 9

Source Open-access PDF More in Software & Coding

importance 4/10 ai assisted Software Testing and Debugging Techniques software engineering stage: research

Problem

Heuristic-based automated program repair wastes resources because mutation operators are chosen uniformly at random, producing many uncompilable or functionality-breaking program variants.

Why it matters

Bug fixing is costly for developers, and wasted search resources slow down automated repair tools.

What the AI did

A reinforcement learning-based approach selects mutation operators during the repair search; it is programming language, granularity-level, and search strategy agnostic and was evaluated empirically against alternative selection techniques.

Results

not stated in source

Impact

The evidence supports feasibility of RL-based operator selection, but the abstract states no measured outcome.

Record

Venue	Automated Software Engineering
Published	2025
Work type	article
Authors	Carol Hanna, Aymeric Blot, Justyna Petke
Citations	9 (OpenAlex, at import time)
References	73
Model or method	not stated in source
Organization	not stated in source
Primary source	https://doi.org/10.1007/s10515-025-00501-z
Open access	hybrid
Keywords	automated program repair, reinforcement learning, mutation operators, empirical evaluation, software engineering

Importance rating

4 / 10

real but narrower advance

Reducing wasted repair search resources helps developers, but abstract reports no noticeable improvement in bugs patched; a narrow methodological advance.

Basis: inferred. Judged from the source text, which gives no explicit scale. Stated scale: not stated in source. Method and limits.

More in Software & Coding

#	Paper	Year	Citations	Importance
1	ImageNet classification with deep convolutional neural networks Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton	2017	106,345	8
2	Deep Residual Learning for Image Recognition Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun	2016	221,134	8
3	Faster sorting algorithms discovered using deep reinforcement learning Daniel J. Mankowitz, Andrea Michi, Anton Zhernov, Marco Gelmi, Marco S...	2023	175	8
4	A deep learning‐based approach for software vulnerability detection using code metrics Fazli Subhan, Xiaoxue Wu, Lili Bo, Xiaobing Sun, Muhammad Rahman	2022	42	7
5	Deep Learning-Based Attack Detection and Classification in Android Devices Alfonso Gómez, Antonio Muñoz	2023	41	7