Security and privacy in federated learning: A survey

Dasaradharami Reddy Kandati* and S Anusha

Federated Learning (FL) allows multiple nodes without actually sharing data with other confidential nodes to retrain a common model. This is particularly relevant in healthcare applications, where data such as medical records are private and confidential. Although federated learning avoids the exchange of actual data, it still remains possible to fight protection on parameter values revealed in the training process or on a generated Machine Learning (ML) model. This study examines FL’s privacy and security concerns and deals with several issues related to privacy protection and safety when developing FL systems. In addition, we have detailed simulation results to illustrate the problems under discussion and potential solutions.

Keywords:

Published on: Aug 16, 2023 Pages: 29-37

Full Text PDF Full Text HTML DOI: 10.17352/tcsit.000066
CrossMark Publons Harvard Library HOLLIS Search IT Semantic Scholar Get Citation Base Search Scilit OAI-PMH ResearchGate Academic Microsoft GrowKudos Universite de Paris UW Libraries SJSU King Library SJSU King Library NUS Library McGill DET KGL BIBLiOTEK JCU Discovery Universidad De Lima WorldCat VU on WorldCat