Ryan Grainger — accepted AI-conference papers

2023

PaCa-ViT: Learning Patch-to-Cluster Attention in Vision Transformers

Ryan Grainger, Thomas Paniagua, Xi Song, Naresh Cuntoor, Mun Wai Lee, Tianfu Wu

CVPR 2023poster

Vision Transformers (ViTs) are built on the assumption of treating image patches as "visual tokens" and learn patch-to-patch attention. The patch embedding based tokenizer has a semantic gap with respect to its counterpart, the textual tokenizer. The patch-to-patch attention suffers from the quadrat…

Cited by 28 Source ↗PDF ↗code ↗Scholar ↗

2023

QuadAttac$K$: A Quadratic Programming Approach to Learning Ordered Top-$K$ Adversarial Attacks

Thomas Paniagua, Ryan Grainger, Tianfu Wu

NeurIPS 2023poster

The adversarial vulnerability of Deep Neural Networks (DNNs) has been well-known and widely concerned, often under the context of learning top-$1$ attacks (e.g., fooling a DNN to classify a cat image as dog). This paper shows that the concern is much more serious by learning significantly more aggre…

Cited by 0 Source ↗PDF ↗Scholar ↗