Artificial Intelligence (AI) is now pervasive in everyday life, and it is quite often based on deep learning techniques. Deep learning continuously proves to be very effective in many applications, but its inherent opacity is also well known: deep learning experts cannot always explain AI decisions, even when coming from systems that were designed by themselves. Thus the need for eXplainable Artificial Intelligence (XAI).
A new paradigm for Artificial Intelligence based on Group Equivariant Non-Expansive Operators (GENEOs) applied to protein pocket detection / G. Bocchi, A. Micheletti, P. Frosini, A. Pedretti, C. Gratteri, F. Lunghini, A.R. Beccari, C. Talarico - In: Proceedings of the Statistics and Data Science Conference / [a cura di] P. Cerchiello, A. Agosto, S. Osmetti, A. Spelta. - Prima edizione. - Pavia : Pavia University Press, 2023 May. - ISBN 9788869521706. - pp. 152-157 (( Intervento presentato al 1. convegno Statistics and Data Science Conference -SDS group Statistics for Data Science and Artificial Intelligence : 27 - 28 April tenutosi a Pavia nel 2023.
A new paradigm for Artificial Intelligence based on Group Equivariant Non-Expansive Operators (GENEOs) applied to protein pocket detection
G. Bocchi
Primo
;A. MichelettiSecondo
;A. Pedretti;
2023
Abstract
Artificial Intelligence (AI) is now pervasive in everyday life, and it is quite often based on deep learning techniques. Deep learning continuously proves to be very effective in many applications, but its inherent opacity is also well known: deep learning experts cannot always explain AI decisions, even when coming from systems that were designed by themselves. Thus the need for eXplainable Artificial Intelligence (XAI).
| File | Dimensione | Formato | |
|---|---|---|---|
|
shortpaper_workshop_sis.pdf
accesso aperto
Descrizione: articolo principale
Tipologia:
Pre-print (manoscritto inviato all'editore)
Dimensione
156.88 kB
Formato
Adobe PDF
|
156.88 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
