Gabriele

Gabriele

PhD student in Explainable AI for Healthcare at the University of Lugano (USI)
Graduated at the University of Cambridge
Ex MIT Research Intern

Publications

Climbing the Ladder of Interpretability with Counterfactual Concept Bottleneck Models

Submitted to IJCAI 2024, 2023

In this paper, we introduce CounterFactual Concept Bottleneck Models (CF-CBMs), a class of models designed to efficiently address three fundamental questions all at once without the need to run post-hoc searches: predict class labels to solve a given classification task (the “What?”), explain task predictions (the “Why?”), and imagine alternative scenarios that could result in different predictions (the “What if?”).

Recommended citation: Gabriele Dominici, Pietro Barbiero, Francesco Giannini, Martin Gjoreski, Marc Langheinrich & Giuseppe Marra. (2024). Climbing the Ladder of Interpretability with Counterfactual Concept Bottleneck Models

SHARCS - Shared Concept Space for Explainable Multimodal Learning

Accepted to NeurIPS2024 Workshop - UniReps, 2023

In this paper, we introduce SHARCS (SHARed Concept Space) – a novel concept-based approach for explainable multimodal learning. SHARCS learns and maps interpretable concepts from different heterogeneous modalities into a single unified concept-manifold, which leads to an intuitive projection of semantically similar cross-modal concepts.

Recommended citation: Gabriele Dominici, Pietro Barbiero, Lucie Charlotte Magister, Pietro Liò, & Nikola Simidjievski. (2023). SHARCS: Shared Concept Space for Explainable Multimodal Learning. https://arxiv.org/abs/2307.00316?context=cs.AI