DeepRNG: Towards Deep Reinforcement Learning-Assisted Generative Testing of Software
This work addresses software testing for developers, but it is incremental as it builds on existing generative testing methods.
The paper tackled the challenge of generative software testing by augmenting random number generators with a deep reinforcement learning agent, resulting in a statistically significant improvement in testing a complex software library with over 350,000 lines of code.
Although machine learning (ML) has been successful in automating various software engineering needs, software testing still remains a highly challenging topic. In this paper, we aim to improve the generative testing of software by directly augmenting the random number generator (RNG) with a deep reinforcement learning (RL) agent using an efficient, automatically extractable state representation of the software under test. Using the Cosmos SDK as the testbed, we show that the proposed DeepRNG framework provides a statistically significant improvement to the testing of the highly complex software library with over 350,000 lines of code. The source code of the DeepRNG framework is publicly available online.