Invited speaker

Ilaria Tiddi

Ilaria Tiddi is an Assistant Professor in Hybrid Intelligence at the Knowledge in AI (KAI) group of the Vrije Universiteit Amsterdam (NL). Her research focuses on creating systems that generate complex narratives through a combination of semantic technologies, open data and machine learning, applied mostly in scientific and robotics scenarios. She is Editor-in-Chief of the CEUR-WS publication, part of the Steering Committee for the Hybrid Human-AI Conference, and Coordinator of the international Staff Exchange for the Dutch Hybrid Intelligence consortium. Since 2014, she is regularly active in the OCs/PCs of the major venues in the KR field (ISWC/ESWC, HHAI, WWW, CIKM, IJCAI/ECAI).

Talk: “Towards Embodied Hybrid Intelligence with Symbolic AI”

Abstract: The Deep Learning revolution has impacted the interpretation of sensor data in autonomous robotics. This success with some costs (limited explainability and trustworthiness, high energy and computational demands, difficulty to handle trivial spatial knowledge) — shortcomings that are particularly problematic for Embodied Hybrid Intelligence (EHI) scenarios where mixed teams of humans and robots closely collaborate to solve tasks. To succeed, EHI systems face 3 challenges: a) robots cannot efficiently represent perceptual knowledge, b) communication in the team lacks common vocabularies, c) accountable solutions to handle failures and unexpected events are missing. In this talk, I will show how these challenges can be addressed as a neuro-symbolic reasoning problem to enable a novel human-robot collaboration, where complementary team members continuously align their goals.