Hydra: Pruning adversarially robust neural networks
In safety-critical but computationally resource-constrained applications, deep learning faces
two key challenges: lack of robustness against adversarial attacks and large neural network …
two key challenges: lack of robustness against adversarial attacks and large neural network …
Advclip: Downstream-agnostic adversarial examples in multimodal contrastive learning
Multimodal contrastive learning aims to train a general-purpose feature extractor, such as
CLIP, on vast amounts of raw, unlabeled paired image-text data. This can greatly benefit …
CLIP, on vast amounts of raw, unlabeled paired image-text data. This can greatly benefit …
Bridging mode connectivity in loss landscapes and adversarial robustness
P Zhao, PY Chen, P Das, KN Ramamurthy… - ar** a robust and constructive framework for
tackling complex learning tasks. Consequently, it is widely utilized in many security-critical …
tackling complex learning tasks. Consequently, it is widely utilized in many security-critical …
Sparsity winning twice: Better robust generalization from more efficient training
Recent studies demonstrate that deep networks, even robustified by the state-of-the-art
adversarial training (AT), still suffer from large robust generalization gaps, in addition to the …
adversarial training (AT), still suffer from large robust generalization gaps, in addition to the …