Virat Shejwalkar (UMass Amherst), Amir Houmansadr (UMass Amherst)

Federated learning (FL) enables many data owners (e.g., mobile devices) to train a joint ML model (e.g., a next-word prediction classifier) without the need of sharing their private training data.

However, FL is known to be susceptible to poisoning attacks by malicious participants (e.g., adversary-owned mobile devices) who aim at hampering the accuracy of the jointly trained model through sending malicious inputs during the federated training process.

In this paper, we present a generic framework for model poisoning attacks on FL. We show that our framework leads to poisoning attacks that substantially outperform state-of-the-art model poisoning attacks by large margins. For instance, our attacks result in $1.5times$ to $60times$ higher reductions in the accuracy of FL models compared to previously discovered poisoning attacks.

Our work demonstrates that existing Byzantine-robust FL algorithms are significantly more susceptible to model poisoning than previously thought. Motivated by this, we design a defense against FL poisoning, called emph{divide-and-conquer} (DnC). We demonstrate that DnC outperforms all existing Byzantine-robust FL algorithms in defeating model poisoning attacks,
specifically, it is $2.5times$ to $12times$ more resilient in our experiments with different datasets and models.

View More Papers

Hunting the Haunter — Efficient Relational Symbolic Execution for...

Lesly-Ann Daniel (CEA, List, France), Sébastien Bardin (CEA, List, France), Tamara Rezk (Inria, France)

Read More

Understanding the Growth and Security Considerations of ECS

Athanasios Kountouras (Georgia Institute of Technology), Panagiotis Kintis (Georgia Institute of Technology), Athanasios Avgetidis (Georgia Institute of Technology), Thomas Papastergiou (Georgia Institute of Technology), Charles Lever (Georgia Institute of Technology), Michalis Polychronakis (Stony Brook University), Manos Antonakakis (Georgia Institute of Technology)

Read More

Awakening the Web's Sleeper Agents: Misusing Service Workers for...

Soroush Karami (University of Illinois at Chicago), Panagiotis Ilia (University of Illinois at Chicago), Jason Polakis (University of Illinois at Chicago)

Read More

Screen Gleaning: A Screen Reading TEMPEST Attack on Mobile...

Zhuoran Liu (Radboud university), Niels Samwel (Radboud University), Léo Weissbart (Radboud University), Zhengyu Zhao (Radboud University), Dirk Lauret (Radboud University), Lejla Batina (Radboud University), Martha Larson (Radboud University)

Read More