报告时间：2026年10月6日下午15:00-16:00

报告地点：育才校区文二楼北楼206广西应用数学中心会议室

报告题目：Explainable anomaly detection using data depth

报告人：帕夫洛·莫扎罗夫斯基（Pavlo Mozharovsky)

报告内容：Anomaly detection is a branch of machine learning and data analysis which aims at identifying observations that exhibit abnormal behaviour. Be it measurement errors, disease development, severe weather, production quality default(s) (items) or failed equipment, financial frauds or crisis events, their on-time identification, isolation and explanation constitute an important task in almost any branch of industry and science. By providing a robust ordering, data depth becomes a particularly useful tool for detection of anomalies, which is even more the case in the unsupervised setting.

In this talk, data depth is studied as an efficient anomaly detection tool, assigning abnormality labels to observations with lower depth values. Practical questions of necessity and reasonability of invariances and shape of the depth function, its robustness and computational complexity, choice of the threshold are discussed. Furthermore, we introduce a new statistical tool dedicated for exploratory analysis of abnormal observations using data depth as a score. Abnormal component analysis (shortly ACA) is a method that searches a low-dimensional data representation that best visualises and explains anomalies. In a comparative simulation and real-data study, ACA proves advantageous for anomaly analysis with respect to methods present in the literature.

报告人简介：Pavlo Mozharovsky is a Full Professor at Institut Polytechnique de Paris (Telecom Paris) in the domain of artificial intelligence. He is the author of major contributions on explainable neural networks, robust and computational statistics, and data depth, and has led a number of open-source implementations. His works include numerous applications in biology, and his career record contains attraction of governmental and industrial funding and successful collaboration with the industry, including strong expertise and delivery of AI-driven solutions over a range of fields.

帕夫洛·莫扎罗夫斯基是巴黎理工学院（电信巴黎）人工智能领域的全职教授。他在可解释神经网络、鲁棒性与计算统计学以及数据深度等领域作出了重要贡献，并主导了多项开源实现。他的研究工作涵盖众多生物学应用，其职业履历包括成功争取政府与产业界资助，并与业界开展卓有成效的合作，尤其在多个领域展现出深厚的专业能力，能够交付基于人工智能的解决方案。