Skip to content

Robust Federated Learning: The Case of Affine Distribution Shifts #43

Description

@nocotan

一言でいうと

Structured affine distribution shiftにおいてロバストなFederated learningのアルゴリズムを提案.

論文リンク

https://papers.nips.cc/paper/2020/file/f5e536083a438cec5b64a4954abc17f1-Paper.pdf

著者/所属機関

Amirhossein Reisizadeh et al.
(ECE Department, UC Santa Barbara)

投稿日付(yyyy/MM/dd)

2020/12

概要

データセットシフト下においてロバストなFederated learningのアルゴリズムを提案.
これを達成するため,ユーザのデータがデバイス依存であるようなstructured affine distribution shiftを考慮する.
このような問題設定においてロバストなFederated Learning framework Robust to Affine distribution shifts (FLRA)を提案.

新規性・差分

  • 提案手法の汎化誤差をPAC-Bayesのフレームワークで理論解析

手法

Screen Shot 2021-02-26 at 3 26 38

結果

Screen Shot 2021-02-26 at 3 26 48

Screen Shot 2021-02-26 at 3 26 53

コメント

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions