Zobrazeno 1 - 10
of 15
pro vyhledávání: '"Bartlomiej Twardowski"'
Publikováno v:
IEEE Access, Vol 12, Pp 45309-45317 (2024)
In this work, we improve the generative replay in a continual learning setting to perform well on challenging scenarios. Because of the growing complexity of continual learning tasks, it is becoming more popular, to apply the generative replay techni
Externí odkaz:
https://doaj.org/article/61daa4fe4f7e4d849ef9c7961da36e3d
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783031282430
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4d78eb608a4d786f51549e24020cdf98
https://doi.org/10.1007/978-3-031-28244-7_4
https://doi.org/10.1007/978-3-031-28244-7_4
Publikováno v:
2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW).
Active learning is a paradigm aimed at reducing the annotation effort by training the model on actively selected informative and/or representative samples. Another paradigm to reduce the annotation effort is self-training that learns from a large amo
Publikováno v:
Fundamenta Informaticae. 168:185-218
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783030721121
ECIR (1)
ECIR (1)
Session-based recommenders, used for making predictions out of users’ uninterrupted sequences of actions, are attractive for many applications. Here, for this task we propose using metric learning, where a common embedding space for sessions and it
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7c9ad3b841c002d9626db755c7d3581b
https://doi.org/10.1007/978-3-030-72113-8_43
https://doi.org/10.1007/978-3-030-72113-8_43
Autor:
Shangling Jui, Xialei Liu, Kai Wang, Joost van de Weijer, Lu Yu, Yongmei Cheng, Luis Herranz, Bartlomiej Twardowski
Publikováno v:
CVPR
Class-incremental learning of deep networks sequentially increases the number of classes to be classified. During training, the network has only access to data of one task at a time, where each task contains several classes. In this setting, networks
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e0a67fd3ba35d88acd4c7748749e107e
http://arxiv.org/abs/2004.00440
http://arxiv.org/abs/2004.00440
Autor:
Marc Masana, Xialei Liu, Bartlomiej Twardowski, Mikel Menta, Andrew D. Bagdanov, Joost van de Weijer
For future learning systems, incremental learning is desirable because it allows for: efficient resource usage by eliminating the need to retrain from scratch at the arrival of new data; reduced memory usage by preventing or limiting the amount of da
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a3dc0498f3df8ab7405639d04e82f70d
Autor:
Joost van de Weijer, Vacit Oguz Yazici, Arnau Ramisa, Bartlomiej Twardowski, Abel Gonzalez-Garcia
Publikováno v:
CVPR
Recurrent neural networks (RNN) are popular for many computer vision tasks, including multi-label classification. Since RNNs produce sequential outputs, labels need to be ordered for the multi-label classification task. Current approaches sort labels
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5c95281c896b1904219b12327c73be97
http://arxiv.org/abs/1911.09996
http://arxiv.org/abs/1911.09996
Autor:
Kacper Radzikowski, Wiktor Kuśmirek, Mateusz Forc, Bartlomiej Twardowski, Łukasz Lepak, Pawel Wawrzynski, Karol Chęciński, Michał Jabłoński, Robert Nowak
Publikováno v:
Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2019.
In this paper we consider the problem of detecting and recognizing widgets in screenshots of computer programs’ graphical user interface (GUI). This problem is fundamental in business process automation. The solution we propose here is based on det
Publikováno v:
Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2018.