BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Privacy-preserving federated learning with multiparty homomorphic 
 Encryption
DTSTART:20221208T160000
DTEND:20221208T170000
DTSTAMP:20260407T091613Z
UID:2ebb8231ea5764a044ddd2f7462a06d2a5a46723d167b13a2b24c854
CATEGORIES:Conferences - Seminars
DESCRIPTION:Sinem SAV\nThe EDIC program is happy to invite you to a public
  talk by our doctoral student Sinem Sav who is doing her PhD in the Securi
 ty and Privacy Engineering Lab - SPRING\nThe aim of the talk is to present
  her achievements to a broad audience to prepare for hiring interviews com
 ing up soon. Be sure to join\, listen to the talk and participate in the Q
 &A session at the end of the presentation.\n\nAbstract\nTraining accurate 
 and robust machine learning models requires a large amount of data that is
  usually scattered across data silos. Sharing or centralizing the data of 
 different healthcare institutions is\, however\, unfeasible or prohibitive
 ly difficult due to privacy regulations. We address the problem of privacy
 -preserving training and evaluation of neural networks in an N-party\, fed
 erated learning setting. Our solutions enable the computation of training 
 under encryption by relying on lattice-based multiparty homomorphic encryp
 tion.\nIn the first part of this talk\, I will present POSEIDON\, the firs
 t of its kind in the regime of privacy-preserving neural network training.
  It preserves the confidentiality of the training data\, the model\, and t
 he evaluation data\, under a passive-adversary model and collusions betwee
 n up to N−1 parties. Then\, I will describe RHODE\, a novel system that 
 enables the training of recurrent neural networks under encryption in the 
 same private federated learning framework. Our experimental results show t
 hat POSEIDON and RHODE achieve accuracy similar to centralized or decentra
 lized non-private approaches and that their computation and communication 
 overhead scales linearly with the number of parties. Finally\, I will demo
 nstrate the applicability of under-encryption training on biomedical analy
 sis for disease-associated cell classification with single-cell analysis. 
 For this\, we design a system\, PriCell\, for training a published state-o
 f-the-art convolutional neural network in a decentralized and privacy-pres
 erving manner. We compare the accuracy achieved by PriCell with the centra
 lized and non-secure solutions and show that PriCell guarantees privacy wi
 thout reducing the utility of the data.\n\nBio\nSinem Sav is a Ph.D. Candi
 date at EPFL\, Switzerland in the groups of Laboratory for Data Security (
 LDS\, 2018-2022) and Security and Privacy Engineering Laboratory (SPRING\,
  2022-2023). She holds a B.Sc. and M.Sc. degree in Computer Engineering fr
 om Bilkent University\, Turkey (2016-2018). Prior to Master’s degree she
  was in Simon Fraser University\, Canada as an undergraduate research assi
 stant.\nSinem’s research is at the intersection of security\, privacy\, 
 and machine learning. She is currently working on privacy-preserving feder
 ated learning by relying on multiparty homomorphic encryption and the appl
 ications of these privacy-preserving systems to biomedical domains. Her wo
 rk on privacy-preserving federated neural network learning is patented and
  received the best paper award on CSAW’21 Applied Research Competition i
 n Europe.\n\n\n 
LOCATION:BC 410 https://plan.epfl.ch/?room==BC%20410
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR