[Computational Learning Theory Team Seminar] Talks on Machine Learning for Mobility and Sensing

We are pleased to announce two research talks by Prof. Rizk Hamada (Osaka Univ.)and Prof. Ahmed Gomaa (EJUST, Egypt) at Kyushu University and online. Their talks are about machine learning for mobility and sensing.

Date: August 18, 2026
Time: 13:30–15:00 (45-minute talk for each, including Q&A)
Online Venue: Open to all registered participants. The seminar will be delivered via Zoom. The URL will be provided only to registered participants.

On-site Venue: Room 313, W2 building, Ito Campus, Kyushu University, Fukuoka

Speaker: Prof. Rizk Hamada (Osaka University)
Title: AI-Driven Spatial Intelligence for Cyber-Physical Systems
Abstract: Spatial intelligence enables cyber-physical systems to infer where people and objects are, how they move, and how environments evolve from incomplete and noisy observations. This talk presents a unified approach to designing AI systems that transform sensing data into spatial representations, predictions, and context-aware actions. A central focus is the learning problem behind these systems: how to achieve strong generalization with limited labeled data, remain robust to device heterogeneity and environmental change, and learn useful representations from large, unstructured spatial data. The talk also examines privacy-preserving human sensing, prediction of complex human behavior, and the integration of explainable reasoning into real-time robotic decision-making. Together, these challenges highlight the connection between spatial intelligence and fundamental questions in sample efficiency, robustness, representation learning, and generalization in real-world cyber-physical systems.

Speaker: Prof. Ahmed Gomaa (EJUST, Egypt)
Title: Deep Learning Revolution in Remote Sensing: From Pixels to Life-Saving Disaster Response

Abstract: Remote sensing poses extreme challenges for computer vision, scale variation, dense objects, high intra-class diversity, and severe domain shift, that cause state-of-the-art deep learning models to fail in unseen environments. This keynote addresses the generalization gap through domain adaptation, multimodal sensor fusion, and edge optimization, arguing that the true frontier lies in building systems that are not just accurate, but generalizable, trustworthy, and deployable in the real world.

We look forward to your participation.