EWHA-SPAI / Journal-Club-2022

Journal Club for 2022
0 stars 0 forks source link

[20220923] Journal Club (PATE, G-PATE) #2

Open capablebear opened 2 years ago

capablebear commented 2 years ago

G-PATE: Scalable Differentially Private Data Generator via Private Aggregation of Teacher Discriminators (NeurIPS 2021)

Motivation

Method

스크린샷 2022-09-23 오후 5 41 03

Contribution

binable43 commented 2 years ago

Semi-supervised Knowledge Transfer for Deep learning from Private Training Data (ICLR 2017)

Motivation

Method

Model Overview 1) Data를 N개의 set(Xn, Yn)으로 분할 -> N개의 dataset에 대하여 N개의 teacher classifier가 각각 훈련 2) Teacher classifier를 이용하여 data label 예측 3) Teacher classifier가 가장 많이 예측한 label을 실제 label로 간주 (=Aggregation) 4) 이를 활용하여 Unseen Public data의 라벨을 예측하는 Student classifier 훈련

Contribution

binable43 commented 2 years ago

[질문 list]