Louise Travé-Massuyès: Contributions of diagnosis to the general demand for AI in the industry
AI applications have never been as popular as today. The enthusiasm of all branches of the industry is in tune and agrees to say that AI technologies can lift many industrial locks for customer value creation, productivity improvement, and insight discovery. Huge business opportunities are expected in this process. Numerous applications are also foreseen in medicine and health care, agriculture and environment, transport/mobility, and energy domains. Faced with this expectation, where does the diagnosis community situate itself?
In this talk, I will focus on engineering and process applications and will identify some requests and needs in these domains. I will then focus on diagnosis reasoning and explain how existing diagnosis theories can bring their contribution based on the presentation of some applications that address specific needs in these domains. I will conclude my talk by drawing my picture of what is still missing to satisfy the current expectations.
Louise Travé-Massuyès holds a position of Directrice de Recherche at Laboratoire d’Analyse et d’Architecture des Systèmes, Centre National de la Recherche Scientifique (LAAS-CNRS), Toulouse, France; head of the Diagnosis and Supervisory Control Team from 1994 to 2015. Her research interests are all related to diagnosis reasoning, tackled by model-based and data-driven approaches. This theme, which she developed throughout her career, led her to consider various formalisms to address a family of problems covered by the diagnosis field. She has been particularly active in establishing bridges between the diagnosis communities of Artificial Intelligence and Automatic Control. She holds a chair on Diagnosis within the Artificial and Natural Intelligence Toulouse Institute ANITI. She is among the coordinators of the “USER” Strategic Field, assigned to diagnosis and health monitoring topics, within the French Aerospace Valley World Competitiveness Cluster, and serves as contact evaluator for the French Research Funding Agency. She serves as Associate Editor for the Artificial Intelligence Journal. She is a member of the International Federation of Automatic Control IFAC Safeprocess Technical Committee.
Meir Kalech: AI for Software Quality Assurance
Modern software systems are highly complex and often have multiple dependencies on external parts such as other processes or services. This poses new challenges and exacerbates existing challenges in different aspects of software Quality Assurance (QA) including testing, debugging and repair. In the last two decades, there is a substantial interest in applying and developing Artificial Intelligence (AI) techniques to address various software engineering challenges. The goal of this talk is to present a novel AI paradigm for software QA (AI4SQ). In particular, we discuss the new challenges that software QA poses, especially those that are due to the rapidly growing number of code revisions. These challenges do not allow the use of traditional manual QA methods and thus there is a need to automate the QA processes. We briefly review state-of-the-art AI techniques to address these challenges and present the AI4SQ paradigm which integrates a range of AI techniques to guide human and computer QA efforts in a cost-effective manner:
A quality assessment AI agent uses machine-learning techniques to predict where coding errors are likely to occur. Then a test generation AI agent considers the error predictions to direct automated test generation. Then a test execution AI agent executes tests, that are passed to the root-cause analysis AI agent, which applies automatic debugging algorithms. The candidate root causes are passed to a code repair AI agent that tries to create a patch for correcting the isolated error.
Meir Kalech completed his Ph.D. at the Computer Science Department of Bar-Ilan University in 2006. In 2008 he became a faculty member of the Department of Software and Information System Engineering at Ben-Gurion University of the Negev. Kalech’s research interests lie in artificial intelligence and specifically in anomaly detection and diagnosis. Kalech established the Anomaly Detection and Diagnosis Lab (AiDnD) which integrates two main AI approaches: model-based and data-driven. He is a recognized expert in model-based diagnosis (MBD) and has published dozens of papers in leading journals and refereed conferences. In the past five years, Kalech promotes research that implements these approaches for software engineering tasks such as debugging and testing. Kalech’s lab promotes cooperation research with the government and leading corporations such as General Motors, Mekorot and IBM. Among the research of Kalech exist anomaly detection of Supervisory Control And Data Acquisition (SCADA) systems, Automated debugging, survival analysis and troubleshooting. Kalech has served as a senior program committee in leading AI conferences such as AAAI, IJCAI, AAMAS and the International Workshop on Principles of Diagnosis.