Zobrazeno 1 - 10
of 46
pro vyhledávání: '"Stefan Holban"'
Autor:
Marina Adriana Mercioni, Stefan Holban
Publikováno v:
2022 International Symposium on Electronics and Telecommunications (ISETC).
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fc0b42a0bd290f65ff01aad1f57f38fa
https://doi.org/10.1109/isetc56213.2022.10010046
https://doi.org/10.1109/isetc56213.2022.10010046
Autor:
Marina-Adriana Mercioni, Stefan Holban
Publikováno v:
2022 IEEE 16th International Symposium on Applied Computational Intelligence and Informatics (SACI).
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7aede1792b763419c8dc7613da8d3d2e
https://doi.org/10.1109/saci55618.2022.9919575
https://doi.org/10.1109/saci55618.2022.9919575
Autor:
Marina Adriana Mercioni, Stefan Holban
Publikováno v:
SACI
This study aims to contribute to the improvement of the network’s performance through developing a novel activation function. Over time, many activation functions have been proposed in order to solve the issues of the previous functions. We note he
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::9651759e2b0609764f49c8ed9b0e214d
https://doi.org/10.1109/saci51354.2021.9465622
https://doi.org/10.1109/saci51354.2021.9465622
Autor:
Marina Adriana Mercioni, Stefan Holban
Publikováno v:
SACI
Our proposal consists of developing two novel activation functions in time series anomaly detection, they have the capability to reduce the validation loss. The approach is based on a current activation function in Deep Learning, a very intensive fie
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5d26638df639f1de897976baba0cb3ca
https://doi.org/10.1109/saci51354.2021.9465604
https://doi.org/10.1109/saci51354.2021.9465604
Autor:
Stefan Holban, Marina Adriana Mercioni
Publikováno v:
ICICT
This study aims to introduce a novel activation function, called Soft Clipping Mish. In other words, it brings improvements in order to increase the performance within the architecture. Its capability was tested on different scenarios using different
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::31384664dec2f3fbece27b01e41d89b6
https://doi.org/10.1109/icict52872.2021.00010
https://doi.org/10.1109/icict52872.2021.00010
Autor:
Stefan Holban, Marina Adriana Mercioni
Publikováno v:
2020 International Symposium on Electronics and Telecommunications (ISETC).
In this paper we proposed two novel activation functions, which we called them TeLU and TeLU learnable. These proposals are a combination of ReLU (Rectified Linear Unit), tangent(tanh), and ELU (Exponential Linear Units) without and with a learnable
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::1695c67bd1dfca20f9f1a0dc76d79097
https://doi.org/10.1109/isetc50328.2020.9301084
https://doi.org/10.1109/isetc50328.2020.9301084
Autor:
Marina Adriana Mercioni, Stefan Holban
Publikováno v:
2020 International Symposium on Electronics and Telecommunications (ISETC).
In order to improve the performance of a deep neural network, the activation function is an important aspect that we must research continuously, that is why we have expanded the research in this direction. We introduced a novel P-Swish activation fun
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7021f5a031dab77fcf4ccbad5f3ee48b
https://doi.org/10.1109/isetc50328.2020.9301059
https://doi.org/10.1109/isetc50328.2020.9301059
Publikováno v:
ICCP
Without activation functions, it would be only possible for the neural network to learn very basic tasks, so the activation function is a key point in the neural network’s architecture. The function allows us to learn more complicated tasks and als
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::48359dc36bcf82dc676cc138ca968142
https://doi.org/10.1109/iccp51029.2020.9266162
https://doi.org/10.1109/iccp51029.2020.9266162
Publikováno v:
Studies in health technology and informatics. 270
In the context of increasing interest in computer-assisted diagnosis for skin lesion images and mobile applications to be used in real life settings, we propose a combined desktop-smartphone solution for dermatological image classification. Hierarchi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=pmid________::b7b9d0272eaa6fcb928deb6a003efa24
https://pubmed.ncbi.nlm.nih.gov/32570630
https://pubmed.ncbi.nlm.nih.gov/32570630
Autor:
Stefan Holban, Marina Adriana Mercioni
Publikováno v:
2020 International Conference on Development and Application Systems (DAS).
This paper is an overview of the most used activation functions, classic functions and current functions as well. When we say classic, we mean the first activation functions, the most popular and used in the past. But due to their disadvantages appea
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::9f2897eefd094011e4899ebacb5a4863
https://doi.org/10.1109/das49615.2020.9108942
https://doi.org/10.1109/das49615.2020.9108942