STRL:基于强化学习的测试算法
STRL:Testing Algorithm Based on Reinforcement Learning
赵花蕊1
作者信息
- 1. 河南省平台经济发展指导中心,河南 郑州 450008
- 折叠
摘要
强化学习因为不需要大量样本进行训练而采用与环境交互的方式产生动态数据的特点,使得强化学习成为近些年机器学习领域的研究热点.本文提出一种新的基于强化学习的软件测试框架STRL,能有效解决回归测试耗时时间久,状态覆盖率低的问题.STRL利用强化学习算法PPO实现高效的自适应探索.实验结果表明,STRL算法在状态覆盖率和测试时间方面都优于人工测试和自动化脚本测试.
Abstract
Reinforcement learning has become research focus in the field of machine learning in recent years due to its character-istic of generating dynamic data through interaction with the environment without requiring a large number of samples for training.This paper proposes a new software testing framework STRL based on reinforcement learning,which can effectively solve the problem of long time consuming and low state coverage of regression testing.STRL utilizes reinforcement learning algorithm PPO to achieve efficient adaptive exploration.Experiments results show that the STRL algorithm outperforms manual testing and auto-mated script testing in terms of state coverage and testing time.
关键词
人工智能软件/智能化软件/传统软件/软件生命周期/强化学习Key words
artificial intelligence software/intelligent software/traditional software/software lifecycle/reinforcement learning引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据