Search Results for author: Athina Markopoulou

Found 7 papers, 2 papers with code

PriPrune: Quantifying and Preserving Privacy in Pruned Federated Learning

no code implementations30 Oct 2023 Tianyue Chu, Mengwei Yang, Nikolaos Laoutaris, Athina Markopoulou

Federated learning (FL) is a paradigm that allows several client devices and a server to collaboratively train a global model, by exchanging only model updates, without the devices sharing their local training data.

Federated Learning

AutoFR: Automated Filter Rule Generation for Adblocking

1 code implementation25 Feb 2022 Hieu Le, Salma Elmalaki, Athina Markopoulou, Zubair Shafiq

AutoFR is effective: it generates filter rules that can block 86% of the ads, as compared to 87% by EasyList, while achieving comparable visual breakage.

Blocking

A Unified Prediction Framework for Signal Maps

no code implementations8 Feb 2022 Emmanouil Alimpertis, Athina Markopoulou, Carter T. Butts, Evita Bakopoulou, Konstantinos Psounis

This improves prediction (e. g., from 64% to 94% in recall for coverage loss) by removing points with negative values, and can also enable data minimization.

Location Leakage in Federated Signal Maps

no code implementations7 Dec 2021 Evita Bakopoulou, Mengwei Yang, Jiang Zhang, Konstantinos Psounis, Athina Markopoulou

We consider the problem of predicting cellular network performance (signal maps) from measurements collected by several mobile devices.

Federated Learning

Walking in Facebook: A Case Study of Unbiased Sampling of OSNs

1 code implementation ‏‏‎ ‎ 2020 Minas Gjoka, Maciej Kurant, Carter T. Butts, Athina Markopoulou

Our goal in this paper is to obtain a representative (unbiased) sample of Facebook users by crawling its social graph.

A Federated Learning Approach for Mobile Packet Classification

no code implementations30 Jul 2019 Evita Bakopoulou, Balint Tillman, Athina Markopoulou

In this paper, we apply, for the first time, a Federated Learning approach to mobile packet classification, which allows mobile devices to collaborate and train a global model, without sharing raw training data.

Classification feature selection +2

PingPong: Packet-Level Signatures for Smart Home Device Events

no code implementations26 Jul 2019 Rahmadi Trimananda, Janus Varmarken, Athina Markopoulou, Brian Demsky

Smart home devices are vulnerable to passive inference attacks based on network traffic, even in the presence of encryption.

Cannot find the paper you are looking for? You can Submit a new open access paper.