In the classical context, the cooperative game theory
concept of the Shapley value has been adapted for post hoc explanations
of Machine Learning models. This approach does not easily translate to
eXplainable Quantum Machine Learning. Finding Shapley values can be
highly computationally complex. In this joint work with Prof. Michel
Barbeau (Carleton University) and Prof. Joaquin Garcia- Alfaro (Télécom
SudParis), we propose a quantum algorithm which can extract Shapley
values within some confidence interval. Our results perform in
polynomial time. We demonstrate the validity of our approach under
specific examples of cooperative voting games.
Bio
Iain Burge is a research assistant at Carleton University
working in the emerging field of quantum explainable AI. He has an
undergraduate degree in cognitive science with a concentration in
computation and minor in mathematics.