Robustness of Meta Matrix Factorization Against Strict Privacy Constraints

Peter Muellner*, Dominik Kowald, Elisabeth Lex

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review

Abstract

In this paper, we explore the reproducibility of MetaMF, a meta matrix factorization framework introduced by Lin et al. MetaMF employs meta learning for federated rating prediction to preserve users’ privacy. We reproduce the experiments of Lin et al. on five datasets, i.e., Douban, Hetrec-MovieLens, MovieLens 1M, Ciao, and Jester. Also, we study the impact of meta learning on the accuracy of MetaMF’s recommendations. Furthermore, in our work, we acknowledge that users may have different tolerances for revealing information about themselves. Hence, in a second strand of experiments, we investigate the robustness of MetaMF against strict privacy constraints. Our study illustrates that we can reproduce most of Lin et al.’s results. Plus, we provide strong evidence that meta learning is essential for MetaMF’s robustness against strict privacy constraints.

Original languageEnglish
Title of host publicationAdvances in Information Retrieval
Subtitle of host publication43rd European Conference on IR Research, ECIR 2021, Proceedings
EditorsDjoerd Hiemstra, Marie-Francine Moens, Josiane Mothe, Raffaele Perego, Martin Potthast, Fabrizio Sebastiani
Place of PublicationCham
PublisherSpringer Science and Business Media Deutschland GmbH
Pages107-119
Number of pages13
Volume2
ISBN (Print)9783030722395
DOIs
Publication statusPublished - 2021
Event43rd European Conference on Information Retrieval: ECIR 2021 - Virtuell
Duration: 28 Mar 20211 Apr 2021

Publication series

NameLecture Notes in Computer Science
Volume12657
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference43rd European Conference on Information Retrieval
Abbreviated titleECIR 2021
CityVirtuell
Period28/03/211/04/21

Keywords

  • Federated learning
  • Matrix factorization
  • Meta learning
  • Privacy
  • Recommender systems
  • Reproducibility

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Robustness of Meta Matrix Factorization Against Strict Privacy Constraints'. Together they form a unique fingerprint.

Cite this