Zobrazeno 1 - 10
of 22
pro vyhledávání: '"Romijnders, Rob"'
The COVID19 pandemic had enormous economic and societal consequences. Contact tracing is an effective way to reduce infection rates by detecting potential virus carriers early. However, this was not generally adopted in the recent pandemic, and priva
Externí odkaz:
http://arxiv.org/abs/2404.13381
The pandemic in 2020 and 2021 had enormous economic and societal consequences, and studies show that contact tracing algorithms can be key in the early containment of the virus. While large strides have been made towards more effective contact tracin
Externí odkaz:
http://arxiv.org/abs/2312.11581
Autor:
Arnab, Anurag, Xiong, Xuehan, Gritsenko, Alexey, Romijnders, Rob, Djolonga, Josip, Dehghani, Mostafa, Sun, Chen, Lučić, Mario, Schmid, Cordelia
Transfer learning is the predominant paradigm for training deep networks on small target datasets. Models are typically pretrained on large ``upstream'' datasets for classification, as such labels are easy to collect, and then finetuned on ``downstre
Externí odkaz:
http://arxiv.org/abs/2207.03807
Autor:
Vasconcelos, Cristina, Larochelle, Hugo, Dumoulin, Vincent, Romijnders, Rob, Roux, Nicolas Le, Goroshin, Ross
We investigate the impact of aliasing on generalization in Deep Convolutional Networks and show that data augmentation schemes alone are unable to prevent it due to structural limitations in widely used architectures. Drawing insights from frequency
Externí odkaz:
http://arxiv.org/abs/2108.03489
Autor:
Minderer, Matthias, Djolonga, Josip, Romijnders, Rob, Hubis, Frances, Zhai, Xiaohua, Houlsby, Neil, Tran, Dustin, Lucic, Mario
Accurate estimation of predictive uncertainty (model calibration) is essential for the safe application of neural networks. Many instances of miscalibration in modern neural networks have been reported, suggesting a trend that newer, more accurate mo
Externí odkaz:
http://arxiv.org/abs/2106.07998
Autor:
Yung, Jessica, Romijnders, Rob, Kolesnikov, Alexander, Beyer, Lucas, Djolonga, Josip, Houlsby, Neil, Gelly, Sylvain, Lucic, Mario, Zhai, Xiaohua
Before deploying machine learning models it is critical to assess their robustness. In the context of deep neural networks for image understanding, changing the object location, rotation and size may affect the predictions in non-trivial ways. In thi
Externí odkaz:
http://arxiv.org/abs/2104.04191
Autor:
Romijnders, Rob, Mahendran, Aravindh, Tschannen, Michael, Djolonga, Josip, Ritter, Marvin, Houlsby, Neil, Lucic, Mario
We propose a method to learn image representations from uncurated videos. We combine a supervised loss from off-the-shelf object detectors and self-supervised losses which naturally arise from the video-shot-frame-object hierarchy present in each vid
Externí odkaz:
http://arxiv.org/abs/2010.02808
Autor:
Djolonga, Josip, Yung, Jessica, Tschannen, Michael, Romijnders, Rob, Beyer, Lucas, Kolesnikov, Alexander, Puigcerver, Joan, Minderer, Matthias, D'Amour, Alexander, Moldovan, Dan, Gelly, Sylvain, Houlsby, Neil, Zhai, Xiaohua, Lucic, Mario
Modern deep convolutional networks (CNNs) are often criticized for not generalizing under distributional shifts. However, several recent breakthroughs in transfer learning suggest that these networks can cope with severe distribution shifts and succe
Externí odkaz:
http://arxiv.org/abs/2007.08558
Training convolutional networks for semantic segmentation with strong (per-pixel) and weak (per-bounding-box) supervision requires a large amount of weakly labeled data. We propose two methods for selecting the most relevant data with weak supervisio
Externí odkaz:
http://arxiv.org/abs/1907.07023
Publikováno v:
IEEE WACV 2019
We propose a normalization layer for unsupervised domain adaption in semantic scene segmentation. Normalization layers are known to improve convergence and generalization and are part of many state-of-the-art fully-convolutional neural networks. We s
Externí odkaz:
http://arxiv.org/abs/1809.05298