Attacking Byzantine Robust Aggregation in High Dimensions

arXiv:2312.14461v2 Announce Type: replace-cross
Abstract: Training modern neural networks or models typically requires averaging over a sample of high-dimensional vectors. Poisoning attacks can skew or bias the average vectors used to train the model, forcing the model to learn specific patterns or avoid learning anything useful. Byzantine robust aggregation is a principled algorithmic defense against such biasing. Robust aggregators can bound the maximum bias in computing centrality statistics, such as mean, even when some fraction of inputs are arbitrarily corrupted. …

abstract aggregation arxiv attacks bias cs.ai cs.cr cs.lg dimensions learn modern networks neural networks patterns poisoning attacks robust sample skew train training type vectors