Zobrazeno 1 - 4
of 4
pro vyhledávání: '"Alexander V. Gayer"'
Publikováno v:
Thirteenth International Conference on Machine Vision.
In this work we study the effect of activation functions in a neural network. We consider how activation functions with different properties and their combination affect the final quality of the model. Due to optimization and speed performance issues
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::79878b0e6d188c0e993eb2fc135ba3ac
https://doi.org/10.1117/12.2587027
https://doi.org/10.1117/12.2587027
Autor:
Alexander V. Gayer, Alexander Sheshkus
Publikováno v:
ICMV
Regularization methods play an important role in artificial neural networks training, improving generalization performance and preventing them from overfitting. In this paper, we introduce a new regularization method, based on the orthogonalization o
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::f08680252cc156c6470dc3834d1b5685
https://doi.org/10.1117/12.2559346
https://doi.org/10.1117/12.2559346
Publikováno v:
ICMV
In this paper we study the real-time augmentation - method of increasing variability of training dataset during the learning process. We consider the most common label-preserving deformations, which can be useful in many practical tasks. Due to limit
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::f8cafa4c0e0453a519a6c6c4af039796
https://doi.org/10.1117/12.2522969
https://doi.org/10.1117/12.2522969
Publikováno v:
ICMV
This paper addresses one of the fundamental problems of machine learning - training data acquiring. Obtaining enough natural training data is rather difficult and expensive. In last years usage of synthetic images has become more beneficial as it all
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::8a4c337832298ce05c49825e704b6abc
https://doi.org/10.1117/12.2310119
https://doi.org/10.1117/12.2310119