Zobrazeno 1 - 10
of 1 107
pro vyhledávání: '"Zarlenga A"'
Autor:
Zarlenga, Mateo Espinosa, Sankaranarayanan, Swami, Andrews, Jerone T. A., Shams, Zohreh, Jamnik, Mateja, Xiang, Alice
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"). Existing bias miti
Externí odkaz:
http://arxiv.org/abs/2409.17691
Concept-based explainability methods provide insight into deep learning systems by constructing explanations using human-understandable concepts. While the literature on human reasoning demonstrates that we exploit relationships between concepts when
Externí odkaz:
http://arxiv.org/abs/2405.18217
Autor:
Dominici, Gabriele, Barbiero, Pietro, Zarlenga, Mateo Espinosa, Termine, Alberto, Gjoreski, Martin, Marra, Giuseppe, Langheinrich, Marc
Causal opacity denotes the difficulty in understanding the "hidden" causal structure underlying the decisions of deep neural network (DNN) models. This leads to the inability to rely on and verify state-of-the-art DNN-based systems, especially in hig
Externí odkaz:
http://arxiv.org/abs/2405.16507
Concept-based methods explain model predictions using human-understandable concepts. These models require accurate concept predictors, yet the faithfulness of existing concept predictors to their underlying concepts is unclear. In this paper, we inve
Externí odkaz:
http://arxiv.org/abs/2401.01259
Publikováno v:
Journal of Dairy Science, Vol 107, Iss 10, Pp 8432-8451 (2024)
ABSTRACT: The aim of this study was to evaluate transcriptional changes in the sole epidermis and dermis of bovine claws with septic sole ulceration of the lateral claw. Assessment included changes in transcripts orchestrating epidermal homeostatic p
Externí odkaz:
https://doaj.org/article/640c62350cf54f9b922b4c64377bc6d2
Publikováno v:
Journal of Dairy Science, Vol 107, Iss 10, Pp 8413-8431 (2024)
ABSTRACT: Pododermatitis aseptica hemorrhagica circumscripta is associated with metalloproteinase 2 weakening of distal phalangeal suspensory structures and sinkage of the distal phalanx in the claw capsule. Pressure from the tuberculum flexorium on
Externí odkaz:
https://doaj.org/article/1dded749c100486bbc8f5a39acdfa073
Autor:
Zarlenga, Mateo Espinosa, Collins, Katherine M., Dvijotham, Krishnamurthy, Weller, Adrian, Shams, Zohreh, Jamnik, Mateja
Concept Bottleneck Models (CBMs) tackle the opacity of neural architectures by constructing and explaining their predictions using a set of high-level concepts. A special property of these models is that they permit concept interventions, wherein use
Externí odkaz:
http://arxiv.org/abs/2309.16928
Autor:
Di Giulio, C., Cardelli, F., Pioli, S., Alesini, D., Bellaveglia, M., Bini, S., Buonomo, B., Cantarella, S., Catuscelli, G., Ceccarelli, M., Ceccarelli, R., Cianfrini, M., Clementi, R., Di Pasquale, E., Di Raddo, G., Di Raddo, R., Falone, A., Gallo, A., Latini, G., Liedl, A., Lollo, V., Piermarini, G., Piersanti, L., Quaglia, S., Rossi, L. A., Sabbatini, L., Scarselletta, G., Scampati, M., Tocci, S., Zarlenga, R.
TEX facility if commissioned for high power testing to characterize accelerating structures and validate them for the operation on future particle accelerators for medical, industrial and research applications. At this aim, TEX is directly involved i
Externí odkaz:
http://arxiv.org/abs/2308.03053
Autor:
Barbiero, Pietro, Ciravegna, Gabriele, Giannini, Francesco, Zarlenga, Mateo Espinosa, Magister, Lucie Charlotte, Tonda, Alberto, Lio', Pietro, Precioso, Frederic, Jamnik, Mateja, Marra, Giuseppe
Publikováno v:
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:1801-1825, 2023
Deep learning methods are highly accurate, yet their opaque decision process prevents them from earning full human trust. Concept-based models aim to address this issue by learning tasks based on a set of human-understandable concepts. However, state
Externí odkaz:
http://arxiv.org/abs/2304.14068
Autor:
Collins, Katherine M., Barker, Matthew, Zarlenga, Mateo Espinosa, Raman, Naveen, Bhatt, Umang, Jamnik, Mateja, Sucholutsky, Ilia, Weller, Adrian, Dvijotham, Krishnamurthy
Placing a human in the loop may abate the risks of deploying AI systems in safety-critical settings (e.g., a clinician working with a medical AI system). However, mitigating risks arising from human error and uncertainty within such human-AI interact
Externí odkaz:
http://arxiv.org/abs/2303.12872