Manish Motwani

Rahil P Mehta, Yashwanthi Anand, Manish Motwani et al.

When an autonomous agent behaves undesirably, including failure to complete a task, it can be difficult to determine whether the behavior is due to a systemic agent error, such as flaws in the model or policy, or an environment error, where a task is inherently infeasible under a given environment configuration, even for an ideal agent. As agents and their environments grow more complex, identifying the error source becomes increasingly difficult but critical for reliable deployment. We introduce AIProbe, a novel black-box testing technique that applies differential testing to attribute undesirable agent behaviors either to agent deficiencies, such as modeling or training flaws, or due to environmental infeasibility. AIProbe first generates diverse environmental configurations and tasks for testing the agent, by modifying configurable parameters using Latin Hypercube sampling. It then solves each generated task using a search-based planner, independent of the agent. By comparing the agent's performance to the planner's solution, AIProbe identifies whether failures are due to errors in the agent's model or policy, or due to unsolvable task conditions. Our evaluation across multiple domains shows that AIProbe significantly outperforms state-of-the-art techniques in detecting both total and unique errors, thereby contributing to a reliable deployment of autonomous agents.

Manish Motwani

Automatic program repair (APR) has recently gained attention because it proposes to fix software defects with no human intervention. To automatically fix defects, most APR tools use the developer-written tests to (a) localize the defect, and (b) generate and validate the automatically produced candidate patches based on the constraints imposed by the tests. While APR tools can produce patches that appear to fix the defect for 11-19% of the defects in real-world software, most of the patches produced are not correct or acceptable to developers because they overfit to the tests used during the repair process. This problem is known as the patch overfitting problem. To address this problem, I propose to equip APR tools with additional constraints derived from natural-language software artifacts such as bug reports and requirements specifications that describe the bug and intended software behavior but are not typically used by the APR tools. I hypothesize that patches produced by APR tools while using such additional constraints would be of higher quality. To test this hypothesis, I propose an automated and objective approach to evaluate the quality of patches and propose two novel methods to improve the fault localization and developer-written test suites using natural-language software artifacts. Finally, I propose to use my patch evaluation methodology to analyze the effect of the improved fault localization and test suites on the quality of patches produced by APR tools for real-world defects.

Manish Motwani, Yuriy Brun

Automated program repair is already deployed in industry, but concerns remain about repair quality. Recent research has shown that one of the main reasons repair tools produce incorrect (but seemingly correct) patches is imperfect fault localization (FL). This paper demonstrates that combining information from natural-language bug reports and test executions when localizing faults can have a significant positive impact on repair quality. For example, existing repair tools with such FL are able to correctly repair 7 defects in the Defects4J benchmark that no prior tools have repaired correctly. We develop, Blues, the first information-retrieval-based, statement-level FL technique that requires no training data. We further develop RAFL, the first unsupervised method for combining multiple FL techniques, which outperforms a supervised method. Using RAFL, we create SBIR by combining Blues with a spectrum-based (SBFL) technique. Evaluated on 815 real-world defects, SBIR consistently ranks buggy statements higher than its underlying techniques. We then modify three state-of-the-art repair tools, Arja, SequenceR, and SimFix, to use SBIR, SBFL, and Blues as their internal FL. We evaluate the quality of the produced patches on 689 real-world defects. Arja and SequenceR significantly benefit from SBIR: Arja using SBIR correctly repairs 28 defects, but only 21 using SBFL, and only 15 using Blues; SequenceR using SBIR correctly repairs 12 defects, but only 10 using SBFL, and only 4 using Blues. SimFix, (which has internal mechanisms to overcome poor FL), correctly repairs 30 defects using SBIR and SBFL, but only 13 using Blues. Our work is the first investigation of simultaneously using multiple software artifacts for automated program repair, and our promising findings suggest future research in this directions is likely to be fruitful.