Follow

RT @j_dreo
This (very new, not peer reviewed) preprint is probably going to be a landmark in ML: "Planting Undetectable Backdoors in Machine Learning Models" arxiv.org/abs/2204.06974
"We show how a malicious [learning service provider] can plant an undetectable backdoor into a classifier"

Sign in to participate in the conversation
Mastodon

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!