Tesi etd-02032025-120342 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
CALABRESE, PIETRO
URN
etd-02032025-120342
Titolo
Analysis of a counterfactual-based feature importance measure: fidelity, computational cost and influencing factors
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
ARTIFICIAL INTELLIGENCE AND DATA ENGINEERING
Relatori
relatore Prof. Cimino, Mario Giovanni Cosimo Antonio
relatore Prof. Alfeo, Antonio Luca
relatore Ing. Gagliardi, Guido
relatore Prof. Alfeo, Antonio Luca
relatore Ing. Gagliardi, Guido
Parole chiave
- counterfactual explanation
- eXplainable Artificial Intelligence
- feature importance
Data inizio appello
21/02/2025
Consultabilità
Non consultabile
Data di rilascio
21/02/2028
Riassunto
In this thesis, a counterfactual-based approach of explainable artificial intelligence is experimented for evaluating its performance in terms of fidelity and computational cost and, as well as, the factors that influence them. The evaluated approach is compared against the state of the art. Experimentation on benchmark datasets testbed is carried out.
File
Nome file | Dimensione |
---|---|
La tesi non è consultabile. |