MUNRO Yann

PhD student at Sorbonne University (ATER, )
Team : LFI
Arrival date : 10/01/2021
    Sorbonne Université - LIP6
    Boîte courrier 169
    Couloir 26-00, Étage 5, Bureau 516
    4 place Jussieu
    75252 PARIS CEDEX 05
    FRANCE

Tel: +33 1 44 27 87 26, Yann.Munro (at) nulllip6.fr
https://lip6.fr/Yann.Munro

Supervision : Isabelle BLOCH

Co-supervision : LESOT Marie-Jeanne

Abstract argumentation and actual causality for the generation of explanations in the context of human-agent interaction

This thesis deals with explainability of artificial intelligence systems, in a context of interactions with humans. The aim is to propose a representation of these interactions that allows the generation of explanations adapted to the explainee. To this end, we have proposed several transformations of an abstract argumentation system into a structural causal model on the one hand, and into an action language on the other. These two logical formalisms make it possible to determine the actual causes of an event, which is the basic building block for generating explanations. We are also working on extending this to the framework of quantitative bipolar argumentation framework, in particular by proposing a new class of gradual semantics.

2022-2024 Publications