University of Tokyo
Creating a quality assurance methodology to ensure trustworthiness
Design and develop systematic quality assurance methodology, technique and toolchains (e.g., testing, analysis, repair) towards approaching the trustworthiness of intelligent systems
Lei Ma
University of Tokyo
Associate Professor
Ongoing Research
- Testing, analysis and repair (etc.) of AI systems from unit level to system level
- Design human interactive and explainable interface for quality assurance of intelligent system
- Building AI-enabled Cyber-Physical Systems and design trustworthiness assurance techniques
- Building AI-enabled Cyber-Cyber Systems and design trustworthiness assurance techniques
- Domain-specific application of designed general-purpose trustworthiness assurance, e.g., autonomous driving, robotics, power system, extensive code, web 3.0, metaverse.
Selected Publications
- ArchRepair: Block-Level Architecture-Oriented Repairing for Deep Neural Networks, ACM Transactions on Software Engineering and Methodology (TOSEM 2023, CORE Rank A*) Media Coverage by NIKKEI News
- SIEGE: A Semantics-Guided Safety Enhancement Framework for AI-Enabled Cyber-Physical Systems, IEEE Transactions on Software Engineering (TSE 2023, CORE Rank A*
- Benchmarking Robustness of AI-enabled Multi-sensor Fusion Systems: Challenges and Opportunities, The 31st the ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE 2023, CORE Rank A*
- DeepSeer: Interactive RNN Explanation and Debugging via State Abstraction, The ACM CHI Conference on Human Factors in Computing Systems (CHI 2023, CORE Rank A*)
- Neural Episodic Control with State Abstraction, The Eleventh International Conference on Learning Representations (ICLR 2023, spotlight CORE Rank A*)
- When Cyber-Physical Systems Meet AI: A Benchmark, an Evaluation, and a Way Forward, The 44th International Conference on Software Engineering (ICSE 2022, CORE Rank A*)
- Automatic RNN Repair via Model-based Analysis, The 38th International Conference on Machine Learning, 2021 (ICML 2021, CORE Rank A*)
- Watch out! Motion is Blurring the Vision of Your Deep Neural Networks, Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS 2020, CORE Rank A*)
- SPARK: Spatial-aware Online Incremental Attack Against Visual Tracking, The 16th European Conference on Computer Vision (ECCV 2020, CORE Rank A*)
- DeepGauge: Multi-Granularity Testing Criteria for Deep Learning Systems, The 33rd IEEE/ACM International Conference on Automated Software Engineering (ASE’18, CORE Rank A*) ACM SIGSOFT Distinguished Paper Award
- GRT: Program-analysis-guided Random Testing, The 30th IEEE/ACM International Conference on Automated Software Engineering. (ASE’15, Core Rank A*) ACM SIGSOFT Distinguished Paper Award