The problem of Byzantine resilience in distributed machine learning, aka Byzantine machine
learning, consists of designing distributed algorithms that can train an accurate model …

This paper addresses the problem of combining Byzantine resilience with privacy in
machine learning (ML). Specifically, we study if a distributed implementation of the …

Privacy-preserving federated learning allows multiple users to jointly train a model with
coordination of a central server. The server only learns the final aggregation result, thereby …

Deep learning (DL) models are enabling a significant paradigm shift in a diverse range of
fields, including natural language processing and computer vision, as well as the design …

Distributed machine learning is primarily motivated by the promise of increased computation
power for accelerating training and mitigating privacy concerns. Unlike machine learning on …

Many fields make use nowadays of machine learning (ML) enhanced applications for cost
optimization, scheduling or forecasting, including the energy sector. However, these very ML …

Algorithms are everywhere. The recipe for the frangipane cake is an algorithm. If all the
listed ingredients are available and the cook is sufficiently deft, after a finite number of small …

We consider the problem of making a multi-agent system (MAS) resilient to Byzantine
failures through replication. We consider a very general model of MAS, where randomness …

In this work we study the problem of Byzantine-robust learning when data among clients is
heterogeneous. We focus on poisoning attacks targeting the convergence of SGD. Although …

The increasing prevalence of personal devices motivates the design of algorithms that can
leverage their computing power, together with the data they generate, in order to build …