Halim Djerroud

I am a researcher in theoretical computer science. My field is automated planning and epistemic reasoning: the design of algorithms that allow an autonomous agent to reason about what it knows, does not know, and can know, and to derive rational decisions in a partially observable, multi-agent setting. The question that structures my contributions is the arbitration between acting, observing, and delegating observation.

My work combines two lines. The first is the design of heuristic search algorithms for epistemic planning, confronted with the community through international competitions; I favour lightweight, legible heuristics over optimal ones, because an inadmissible yet informative heuristic often reveals the structure of a problem better than an exhaustive computation. The second is more theoretical: the study of the structural conditions under which classical completeness guarantees can be transposed to the multi-agent epistemic setting, by analysing properties of the interaction protocol rather than modifying the formalism.

Theory, however, is not enough. Just as a phenomenon predicted in theoretical physics is only established once it has been observed, my algorithms and theorems must be confronted with the real world. I therefore design and build the experimental apparatus that allows this observation: an instrumented physical environment on which autonomous agents, instantiating my planners, act in the real world. The research page develops this approach further.