Explainable optimization

Optimization solvers are often unexplainable black boxes where the reasoning for a solution is inaccessible to users. We propose abstract argumentation, a subfield of artificial intelligence, to empower the interaction between optimization solvers and users, supported by tractable explanations which certify or refute solutions (Čyras et al., 2019, Čyras et al, 2020).

Our approach consists of hooking an optimization solver to the back-end of an argumentation framework and then letting the argumentation framework interact with the user.

This video introduces our work, which won Best (Innovative) Demonstration at AAMAS 2020:

Collaborators: Dr Kristijonas Čyras, Dr Dimitrios Letsios, & Prof Francesca Toni

Delicious Twitter Digg this StumbleUpon Facebook