Solving weakly supervised multi-output regression Conference Abstracts
| Conference |
Международная конференция КОМПЬЮТЕРНЫЕ ДАННЫЕ
АНАЛИЗ И МОДЕЛИРОВАНИЕ: СТОХАСТИКА И
НАУКА ДАННЫХ 06-10 Sep 2022 , Минск |
||||
|---|---|---|---|---|---|
| Source | Computer data analysis and modeling: Stochastics and data science : Proceedings of the XIII International Conference. Minsk, 2022 Compilation, Belarusian State University. Minsk.2022. 250 c. ISBN 978-985-881-420-5. |
||||
| Output data | Year: 2022, Pages: 83-86 Pages count : 4 | ||||
| Tags | data science, supervised regression, wasserstein distance | ||||
| Authors |
|
||||
| Affiliations |
|
Funding (1)
| 1 | Russian Science Foundation | 22-21-00261 |
Abstract:
We propose a solution to the multi-output weakly supervised regression problem. In the studied setting the observed data is partly labeled, and known labels are considered to be the probability distribution to represent possible uncertainty in labeling due to noise. The proposed solution consists in minimizing the Wasser-stein distance between multivariate normal distributions, and approximation of matrices having a low-rank format. In the experimental part of the paper we provide the results, which are shown to be superior to the previous methods on Monte-Carlo simulations and a real dataset.
Cite:
Kondratyev V.
, Berikov V.
Solving weakly supervised multi-output regression
In compilation Computer data analysis and modeling: Stochastics and data science : Proceedings of the XIII International Conference. Minsk, 2022. – Belarusian State University., 2022. – C.83-86. – ISBN 978-985-881-420-5. РИНЦ
Solving weakly supervised multi-output regression
In compilation Computer data analysis and modeling: Stochastics and data science : Proceedings of the XIII International Conference. Minsk, 2022. – Belarusian State University., 2022. – C.83-86. – ISBN 978-985-881-420-5. РИНЦ
Identifiers:
| Elibrary: | 65639644 |
Citing:
Пока нет цитирований