Zobrazeno 1 - 10
of 89
pro vyhledávání: '"Sander, Tom"'
Image watermarking methods are not tailored to handle small watermarked areas. This restricts applications in real-world scenarios where parts of the image may come from different sources or have been edited. We introduce a deep-learning model for lo
Externí odkaz:
http://arxiv.org/abs/2411.07231
Autor:
Labiad, Ismail, Bäck, Thomas, Fernandez, Pierre, Najman, Laurent, Sander, Tom, Ye, Furong, Zameshina, Mariia, Teytaud, Olivier
In many cases, adversarial attacks are based on specialized algorithms specifically dedicated to attacking automatic image classifiers. These algorithms perform well, thanks to an excellent ad hoc distribution of initial attacks. However, these attac
Externí odkaz:
http://arxiv.org/abs/2409.15119
Autor:
Sander, Tom, Yu, Yaodong, Sanjabi, Maziar, Durmus, Alain, Ma, Yi, Chaudhuri, Kamalika, Guo, Chuan
Differentially private (DP) machine learning is considered the gold-standard solution for training a model from sensitive data while still preserving privacy. However, a major barrier to achieving this ideal is its sub-optimal privacy-accuracy trade-
Externí odkaz:
http://arxiv.org/abs/2403.02506
We investigate the radioactivity of text generated by large language models (LLM), i.e. whether it is possible to detect that such synthetic input was used to train a subsequent LLM. Current methods like membership inference or active IP protection e
Externí odkaz:
http://arxiv.org/abs/2402.14904
Training Deep Neural Networks (DNNs) with small batches using Stochastic Gradient Descent (SGD) yields superior test performance compared to larger batches. The specific noise structure inherent to SGD is known to be responsible for this implicit bia
Externí odkaz:
http://arxiv.org/abs/2402.08344
Differentially Private methods for training Deep Neural Networks (DNNs) have progressed recently, in particular with the use of massive batches and aggregated data augmentations for a large number of training steps. These techniques require much more
Externí odkaz:
http://arxiv.org/abs/2210.03403
Autor:
Sloka, Biruta, Sander, Tom
Publikováno v:
European Integration Studies. 13(1):18-36
Externí odkaz:
https://www.ceeol.com/search/article-detail?id=818248
Publikováno v:
International Journal of Web Information Systems, 2017, Vol. 13, Issue 1, pp. 14-24.
Externí odkaz:
http://www.emeraldinsight.com/doi/10.1108/IJWIS-06-2016-0029
Publikováno v:
Economics and Business, Vol 30, Iss 1, Pp 51-61 (2017)
The research investigates the question of the importance of mentioning the name of the employer in online and offline social networks. The exchange of information is an important part of social networks and social capital theory. Companies can use th
Externí odkaz:
https://doaj.org/article/b543b60d27934067bd0deaac791e3956
Publikováno v:
Business, Management and Education. 14(2):179-193
Externí odkaz:
https://www.ceeol.com/search/article-detail?id=475700