no code implementations • 26 Sep 2024 • Mateo Espinosa Zarlenga, Swami Sankaranarayanan, Jerone T. A. Andrews, Zohreh Shams, Mateja Jamnik, Alice Xiang
Deep neural networks trained via empirical risk minimisation often exhibit significant performance disparities across groups, particularly when group and task labels are spuriously correlated (e. g., "grassy background" and "cows").
1 code implementation • 28 May 2024 • Naveen Raman, Mateo Espinosa Zarlenga, Mateja Jamnik
Concept-based explainability methods provide insight into deep learning systems by constructing explanations using human-understandable concepts.
no code implementations • 26 May 2024 • Gabriele Dominici, Pietro Barbiero, Mateo Espinosa Zarlenga, Alberto Termine, Martin Gjoreski, Giuseppe Marra, Marc Langheinrich
Causal opacity denotes the difficulty in understanding the "hidden" causal structure underlying the decisions of deep neural network (DNN) models.
1 code implementation • 2 Jan 2024 • Naveen Raman, Mateo Espinosa Zarlenga, Juyeon Heo, Mateja Jamnik
These models require accurate concept predictors, yet the faithfulness of existing concept predictors to their underlying concepts is unclear.
1 code implementation • NeurIPS 2023 • Mateo Espinosa Zarlenga, Katherine M. Collins, Krishnamurthy Dvijotham, Adrian Weller, Zohreh Shams, Mateja Jamnik
To address this, we propose Intervention-aware Concept Embedding models (IntCEMs), a novel CBM-based architecture and training paradigm that improves a model's receptiveness to test-time interventions.
1 code implementation • 27 Apr 2023 • Pietro Barbiero, Gabriele Ciravegna, Francesco Giannini, Mateo Espinosa Zarlenga, Lucie Charlotte Magister, Alberto Tonda, Pietro Lio', Frederic Precioso, Mateja Jamnik, Giuseppe Marra
Deep learning methods are highly accurate, yet their opaque decision process prevents them from earning full human trust.
no code implementations • 22 Mar 2023 • Katherine M. Collins, Matthew Barker, Mateo Espinosa Zarlenga, Naveen Raman, Umang Bhatt, Mateja Jamnik, Ilia Sucholutsky, Adrian Weller, Krishnamurthy Dvijotham
We study how existing concept-based models deal with uncertain interventions from humans using two novel datasets: UMNIST, a visual dataset with controlled simulated uncertainty based on the MNIST dataset, and CUB-S, a relabeling of the popular CUB concept dataset with rich, densely-annotated soft labels from humans.
1 code implementation • 25 Jan 2023 • Mateo Espinosa Zarlenga, Pietro Barbiero, Zohreh Shams, Dmitry Kazhdan, Umang Bhatt, Adrian Weller, Mateja Jamnik
In this paper, we show that such metrics are not appropriate for concept learning and propose novel metrics for evaluating the purity of concept representations in both approaches.
1 code implementation • 19 Sep 2022 • Mateo Espinosa Zarlenga, Pietro Barbiero, Gabriele Ciravegna, Giuseppe Marra, Francesco Giannini, Michelangelo Diligenti, Zohreh Shams, Frederic Precioso, Stefano Melacci, Adrian Weller, Pietro Lio, Mateja Jamnik
Deploying AI-powered systems requires trustworthy models supporting effective human interactions, going beyond raw prediction accuracy.
1 code implementation • 24 Nov 2021 • Mateo Espinosa Zarlenga, Zohreh Shams, Mateja Jamnik
In recent years, there has been significant work on increasing both interpretability and debuggability of a Deep Neural Network (DNN) by extracting a rule-based model that approximates its decision boundary.
no code implementations • 29 Sep 2021 • Mateo Espinosa Zarlenga, Pietro Barbiero, Zohreh Shams, Dmitry Kazhdan, Umang Bhatt, Mateja Jamnik
Recent work on Explainable AI has focused on concept-based explanations, where deep learning models are explained in terms of high-level units of information, referred to as concepts.