Tesi etd-09182021-111639 |
Link copiato negli appunti
Tipo di tesi
Tesi di laurea magistrale
Autore
FONTANA, MICHELE
URN
etd-09182021-111639
Titolo
A new approach for cross-silo federated learning and its privacy risks
Dipartimento
INFORMATICA
Corso di studi
INFORMATICA
Relatori
relatore Prof.ssa Monreale, Anna
relatore Dott.ssa Naretto, Francesca
relatore Dott.ssa Naretto, Francesca
Parole chiave
- privacy risk assessment
- federated learning
Data inizio appello
08/10/2021
Consultabilità
Non consultabile
Data di rilascio
08/10/2024
Riassunto
In this thesis, we define a novel federated learning approach tailored for training machine learning models on data distributed over federated organizations hierarchically organized.
Then, we also tackle the problem of assessing the privacy risk of users represented in the training data. We empirically show, by attacking the federated models with the Membership Inference Attack, that the privacy of the users in the training data may have high risk.
Then, we also tackle the problem of assessing the privacy risk of users represented in the training data. We empirically show, by attacking the federated models with the Membership Inference Attack, that the privacy of the users in the training data may have high risk.
File
| Nome file | Dimensione |
|---|---|
La tesi non è consultabile. |
|