Invited Talk
in
Workshop: Generalization for Time Series in Tight Settings: Latency, Inference, Memory, prIvacy and Sustainability (TS-LIMITS) Sun, Dec 13, 2026 • 7:10 PM – 7:50 PM AEDT

Keynote: João Gama (University of Porto) “A Neuro-symbolic approach for explaining Rare Events”

Joao Gama

Abstract

In this talk, we describe a neural-symbolic architecture for explaining rare events. We propose a two-layer system, where the first layer is unsupervised, based on autoencoders, and is designed to detect rare events (outliers, drifts, anomalies, failures). The second layer is supervised, rule-based, and used to explain the anomalies detected in the first layer. Both models run online and in parallel. We evaluate the proposed system in a real-world predictive maintenance case study, presenting examples of explanations that illustrate their benefits.

Speaker

Joao Gama

João Gama is an Emeritus Professor at the University of Porto, Portugal. He received his Ph.D. in Computer Science from the University of Porto in 2000. He taught Informatics and data sciences at the School of Economics and Management - UP for more than 30 years. He was Director of the Master in Data Analytics for 12 years. He is EurAI Fellow, IEEE Fellow, and Fellow of the Asia-Pacific AI Association. He is member of the Academia das Ciências de Lisboa. He is member of the board of directors of the LIAAD, a group belonging to INESC Tec. His main scientific contributions are in the area of learning from data streams, where he has an extensive list of publications. He is the Editor-in-Chief of the International Journal of Data Science and Analytics, published by Springer.

Chat is not available.