In the rapidly evolving landscape of AI, data-driven approaches have emerged as a dominant trend. From a planning and scheduling perspective, this trend is exemplified by two crucial technical artifacts gaining prominence. The first are planning models that are (partially) learned from data (e.g., a weather forecast in a model of flight actions). The second are action-decision components learned from data, in particular, action policies or planning-control knowledge for making decisions in dynamic environments (e.g., manufacturing processes under resource-availability and job-length fluctuations).

Given the nature of such data-driven artifacts, reliability is a key concern, prominently including safety, robustness, and fairness in various forms, but possibly other concerns as well. Arguably, this is one of the grand challenges in AI for the foreseeable future.

Several TUPLES members have been involved in the organizing committee: Jesse Davis of KU Leuven, Daniel Höller and Jörg Hoffmann of Saarland University, Michele Lombardi of DISI University of Bologna and the Coordinator of the TUPLES Project Sylvie Thiebaux of the University of Toulouse.

Among the accepted papers two were related to the TUPLES’ project:

– Xandra Schuler, Jan Eisenhut, Daniel Höller, Daniel Fišer and Joerg Hoffmann: Action Policy Testing with Heuristic-Based Bias Functions – pdf

– Marcel Vinzent, Min Wu, Haoze Wu and Joerg Hoffmann: Policy-Specific Abstraction Predicate Selection in Neural Policy Safety Verification pdf

For details on the workshop:

https://icaps23.icaps-conference.org/program/workshops/rddps/