Generating test input with deep reinforcement learning

Test data generation is a tedious and laborious process. Search-based Software Testing (SBST) automatically generates test data optimising structural test criteria using metaheuristic algorithms. In essence, metaheuristic algorithms are systematic trial-and-error based on the feedback of fitness fun...

Full description

Saved in:
Bibliographic Details
Published in:2018 IEEE ACM 11th International Workshop on Search Based Software Testing (SBST) pp. 51 - 58
Main Authors: Kim, Junhwi, Kwon, Minhyuk, Yoo, Shin
Format: Conference Proceeding
Language:English
Published: New York, NY, USA ACM 28.05.2018
Series:ACM Conferences
Subjects:
Instruments
Learning (artificial intelligence)
Machine learning
Neural networks
Python
reinforcement learning
sbst
Software algorithms
Software and its engineering > Software creation and management > Search-based software engineering
test data generation
ISBN:9781450357418, 1450357415
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Test data generation is a tedious and laborious process. Search-based Software Testing (SBST) automatically generates test data optimising structural test criteria using metaheuristic algorithms. In essence, metaheuristic algorithms are systematic trial-and-error based on the feedback of fitness function. This is similar to an agent of reinforcement learning which iteratively decides an action based on the current state to maximise the cumulative reward. Inspired by this analogy, this paper investigates the feasibility of employing reinforcement learning in SBST to replace human designed meta-heuristic algorithms. We reformulate the software under test (SUT) as an environment of reinforcement learning. At the same time, we present GunPowder, a novel framework for SBST which extends SUT to the environment. We train a Double Deep Q-Networks (DDQN) agent with deep neural network and evaluate the effectiveness of our approach by conducting a small empirical study. Finally, we find that agents can learn metaheuristic algorithms for SBST, achieving 100% branch coverage for training functions. Our study sheds light on the future integration of deep neural network and SBST.
ISBN:9781450357418
1450357415
DOI:10.1145/3194718.3194720