Litcius/Paper detail

Federated learning: Applications, Security hazards and Defense measures

Sonam Tyagi, Ishwari Singh Rajput, Richa Pandey

202336 citationsDOI

Abstract

Federated learning(FL), a cutting-edge method of distributed learning, enables multiple users to share training results while maintaining the privacy of their personal data. Collecting data from different data owners for making machine learning predictions becomes increasingly challenging as data security becomes more of a priority. Federated learning protects user’s privacy in addition to increase the training data while overcoming the challenges faced by machine learning and deep learning models. Since the data privacy and security is a world-wide concern, the concept of federated learning is increasing day by day from theoretical to practical level. This review paper involves the overview of the federated learning framework, its types, different applications, several types of attacks and defense mechanism.

Topics & Concepts

Computer scienceFederated learningInformation privacyComputer securityMechanism (biology)Enhanced Data Rates for GSM EvolutionDeep learningData securityArtificial intelligenceMachine learningData sciencePhilosophyEncryptionEpistemologyPrivacy-Preserving Technologies in DataInternet Traffic Analysis and Secure E-votingPrivacy, Security, and Data Protection