Zobrazeno 1 - 7
of 7
pro vyhledávání: '"Ramsauer, Hubert"'
One of the main tasks of an autonomous agent in a vehicle is to correctly perceive its environment. Much of the data that needs to be processed is collected by optical sensors such as cameras. Unfortunately, the data collected in this way can be affe
Externí odkaz:
http://arxiv.org/abs/2305.12983
Autor:
Fürst, Andreas, Rumetshofer, Elisabeth, Lehner, Johannes, Tran, Viet, Tang, Fei, Ramsauer, Hubert, Kreil, David, Kopp, Michael, Klambauer, Günter, Bitto-Nemling, Angela, Hochreiter, Sepp
CLIP yielded impressive results on zero-shot transfer learning tasks and is considered as a foundation model like BERT or GPT3. CLIP vision models that have a rich representation are pre-trained using the InfoNCE objective and natural language superv
Externí odkaz:
http://arxiv.org/abs/2110.11316
Autor:
Widrich, Michael, Schäfl, Bernhard, Ramsauer, Hubert, Pavlović, Milena, Gruber, Lukas, Holzleitner, Markus, Brandstetter, Johannes, Sandve, Geir Kjetil, Greiff, Victor, Hochreiter, Sepp, Klambauer, Günter
Publikováno v:
Advances in Neural Information Processing Systems 33 (NeurIPS 2020)
A central mechanism in machine learning is to identify, store, and recognize patterns. How to learn, access, and retrieve such patterns is crucial in Hopfield networks and the more recent transformer architectures. We show that the attention mechanis
Externí odkaz:
http://arxiv.org/abs/2007.13505
Autor:
Ramsauer, Hubert, Schäfl, Bernhard, Lehner, Johannes, Seidl, Philipp, Widrich, Michael, Adler, Thomas, Gruber, Lukas, Holzleitner, Markus, Pavlović, Milena, Sandve, Geir Kjetil, Greiff, Victor, Kreil, David, Kopp, Michael, Klambauer, Günter, Brandstetter, Johannes, Hochreiter, Sepp
We introduce a modern Hopfield network with continuous states and a corresponding update rule. The new Hopfield network can store exponentially (with the dimension of the associative space) many patterns, retrieves the pattern with one update, and ha
Externí odkaz:
http://arxiv.org/abs/2008.02217
Autor:
Unterthiner, Thomas, Nessler, Bernhard, Seward, Calvin, Klambauer, Günter, Heusel, Martin, Ramsauer, Hubert, Hochreiter, Sepp
Generative adversarial networks (GANs) evolved into one of the most successful unsupervised techniques for generating realistic images. Even though it has recently been shown that GAN training converges, GAN models often end up in local Nash equilibr
Externí odkaz:
http://arxiv.org/abs/1708.08819
Publikováno v:
Advances in Neural Information Processing Systems 30 (NIPS 2017)
Generative Adversarial Networks (GANs) excel at creating realistic images with complex models for which maximum likelihood is infeasible. However, the convergence of GAN training has still not been proved. We propose a two time-scale update rule (TTU
Externí odkaz:
http://arxiv.org/abs/1706.08500
Autor:
Ramsauer, Hubert
submitted by Hubert Ramsauer Universität Linz, Univ., Masterarbeit, 2017
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od______3361::d77b5600a4d6f62a0f26c0058e315871