all AI news
LGV: Boosting Adversarial Example Transferability from Large Geometric Vicinity. (arXiv:2207.13129v1 [cs.LG])
July 28, 2022, 1:11 a.m. | Martin Gubri, Maxime Cordy, Mike Papadakis, Yves Le Traon, Koushik Sen
stat.ML updates on arXiv.org arxiv.org
We propose transferability from Large Geometric Vicinity (LGV), a new
technique to increase the transferability of black-box adversarial attacks. LGV
starts from a pretrained surrogate model and collects multiple weight sets from
a few additional training epochs with a constant and high learning rate. LGV
exploits two geometric properties that we relate to transferability. First,
models that belong to a wider weight optimum are better surrogates. Second, we
identify a subspace able to generate an effective surrogate ensemble among this …
More from arxiv.org / stat.ML updates on arXiv.org
Jobs in AI, ML, Big Data
Lead GNSS Data Scientist
@ Lurra Systems | Melbourne
Senior Machine Learning Engineer (MLOps)
@ Promaton | Remote, Europe
Data Analyst
@ Aviva | UK - Norwich - Carrara - 1st Floor
Werkstudent im Bereich Performance Engineering mit Computer Vision (w/m/div.) - anteilig remote
@ Bosch Group | Stuttgart, Lollar, Germany
Applied Research Scientist - NLP (Senior)
@ Snorkel AI | Hybrid / San Francisco, CA
Associate Principal Engineer, Machine Learning
@ Nagarro | Remote, India