2 code implementations • • Micah Goldblum, Hossein Souri, Renkun Ni, Manli Shu, Viraj Prabhu, Gowthami Somepalli, Prithvijit Chattopadhyay, Mark Ibrahim, Adrien Bardes, Judy Hoffman, Rama Chellappa, Andrew Gordon Wilson, Tom Goldstein
Battle of the Backbones (BoB) makes this choice easier by benchmarking a diverse suite of pretrained models, including vision-language models, those trained via self-supervised learning, and the Stable Diffusion backbone, across a diverse set of computer vision tasks ranging from classification to object detection to OOD generalization and more.
Federated learning is particularly susceptible to model poisoning and backdoor attacks because individual users have direct control over the training data and model updates.
Deep learning is increasingly moving towards a transfer learning paradigm whereby large foundation models are fine-tuned on downstream tasks, starting from an initialization learned on the source task.
This is an unfortunate omission, as 'imbalance' is a more complex matter in identification; imbalance may arise in not only the training data, but also the testing data, and furthermore may affect the proportion of identities belonging to each demographic group or the number of images belonging to each identity.
Under JSTM, we develop novel adversarial attacks and defenses.
In other words, we can make a weak model more robust with the help of a strong teacher model.
The adversarial attack literature contains a myriad of algorithms for crafting perturbations which yield pathological behavior in neural networks.
In contrast, the Hidden Trigger Backdoor Attack achieves poisoning without placing a trigger into the training data at all.
Watermarking is a commonly used strategy to protect creators' rights to digital images, videos and audio.
To tackle this issue, we take an information-theoretic approach and maximize a variational lower bound on the entropy of the generated samples to increase their diversity.
Therefore, we present a novel `Adversarial Gender De-biasing algorithm (AGENDA)' to reduce the gender information present in face descriptors obtained from previously trained face recognition networks.
To mitigate the degradation due to turbulence which includes deformation and blur, we propose a generative single frame restoration algorithm which disentangles the blur and deformation due to turbulence and reconstructs a restored image.