no code implementations • 21 Jan 2024 • Siddharth Mansingh, Michal Kucer, Garrett Kenyon, Juston Moore, Michael Teti
Deep neural networks (DNNs) are easily fooled by adversarial perturbations that are imperceptible to humans.
1 code implementation • 7 Nov 2023 • Kehinde Ajayi, Xin Wei, Martin Gryder, Winston Shields, Jian Wu, Shawn M. Jones, Michal Kucer, Diane Oyen
CV tasks, such as image captioning, which has primarily been carried out on natural images, still struggle to produce accurate and meaningful captions on sketched images often included in scientific and technical documents.
no code implementations • 2 Jun 2022 • Diane Oyen, Michal Kucer, Nick Hengartner, Har Simrat Singh
However, for the special case of class-dependent label noise (independent of features given the class label), the tipping point can be as low as 50%.
1 code implementation • IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2022 • Michal Kucer, Diane Oyen, Juan Castorena
First, we introduce DeepPatent, a new large-scale dataset for recognition and retrieval of design patent drawings.
Ranked #3 on Image Retrieval on DeepPatent
no code implementations • 8 Dec 2021 • Michal Kucer, Diane Oyen, Garrett Kenyon
We identify primary ways in which self-supervision can be added to adversarial training, and observe that using a self-supervised loss to optimize both network parameters and find adversarial examples leads to the strongest improvement in model robustness, as this can be viewed as a form of ensemble adversarial training.
no code implementations • NeurIPS Workshop ImageNet_PPF 2021 • Michal Kucer, Diane Oyen
Though much previous work tried to uncover the best practices for transfer learning, much is left unexplored.