Please note: This PhD seminar will take place in DC 2564 and online.
Mengxiao (Max) Zhang, PhD candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Chengnian Sun
Given a list L of elements and a property ψ that L exhibits, ddmin is a classic test input minimization algorithm that aims to automatically remove ψ-irrelevant elements from L. This algorithm has been widely adopted in domains such as test input minimization and software debloating. Recently, ProbDD, a variant of ddmin, has been proposed and achieved state-of-the-art performance. By employing Bayesian optimization, ProbDD estimates the probability of each element in L being relevant to ψ, and statistically decides which and how many elements should be deleted together each time. However, the theoretical probabilistic model of ProbDD is rather intricate, and the underlying details for the superior performance of ProbDD have not been adequately explored.
In this paper, we conduct the first in-depth theoretical analysis of ProbDD, clarifying the trends in probability and subset size changes and simplifying the probability model. We complement this analysis with empirical experiments, including success rate analysis, ablation studies, and examinations of trade-offs and limitations, to further comprehend and demystify this state-of-the-art algorithm. Our success rate analysis reveals how ProbDD effectively addresses bottlenecks that slow down ddmin by skipping inefficient queries that attempt to delete complements of subsets and previously tried subsets. The ablation study illustrates that randomness in ProbDD has no significant impact on efficiency. These findings provide valuable insights for future research and applications of test input minimization algorithms.
Based on the findings above, we propose CDD, a simplified version of ProbDD, reducing the complexity in both theory and implementation. CDD assists in 1 validating the correctness of our key findings, e.g., that probabilities in ProbDD essentially serve as monotonically increasing counters for each element, and 2 identifying the main factors that truly contribute to ProbDD’s superior performance. Our comprehensive evaluations across 76 benchmarks in test input minimization and software debloating demonstrate that CDD can achieve the same performance as ProbDD, despite being much simplified.
To attend this PhD seminar in person, please go to DC 2564. You can also attend virtually using MS Teams.