Geodesic
Loss functions and descent method
Proceedings of the Yerevan State University. Physical and mathematical sciences, Tome 55 (2021) no. 1, pp. 29-35.

Voir la notice de l'article provenant de la source Math-Net.Ru

In this paper, we showed that it is possible to use gradient descent method to get minimal error values of loss functions close to their Bayesian estimators. We calculated Bayesian estimators mathematically for different loss functions and tested them using gradient descent algorithm. This algorithm, working on Normal and Poisson distributions showed that it is possible to find minimal error values without having Bayesian estimators. Using Python, we tested the theory on loss functions with known Bayesian estimators as well as another loss functions, getting results proving the theory.
Keywords: Bayesian estimators, loss functions, machine learning.
Mots-clés : gradient descent 
@article{UZERU_2021_55_1_a3,
     author = {V. K. Ohanyan and H. Z. Zohrabyan},
     title = {Loss functions and descent method},
     journal = {Proceedings of the Yerevan State University. Physical and mathematical sciences},
     pages = {29--35},
     publisher = {mathdoc},
     volume = {55},
     number = {1},
     year = {2021},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/UZERU_2021_55_1_a3/}
}
TY  - JOUR
AU  - V. K. Ohanyan
AU  - H. Z. Zohrabyan
TI  - Loss functions and descent method
JO  - Proceedings of the Yerevan State University. Physical and mathematical sciences
PY  - 2021
SP  - 29
EP  - 35
VL  - 55
IS  - 1
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/UZERU_2021_55_1_a3/
LA  - en
ID  - UZERU_2021_55_1_a3
ER  - 
%0 Journal Article
%A V. K. Ohanyan
%A H. Z. Zohrabyan
%T Loss functions and descent method
%J Proceedings of the Yerevan State University. Physical and mathematical sciences
%D 2021
%P 29-35
%V 55
%N 1
%I mathdoc
%U http://geodesic.mathdoc.fr/item/UZERU_2021_55_1_a3/
%G en
%F UZERU_2021_55_1_a3
V. K. Ohanyan; H. Z. Zohrabyan. Loss functions and descent method. Proceedings of the Yerevan State University. Physical and mathematical sciences, Tome 55 (2021) no. 1, pp. 29-35. http://geodesic.mathdoc.fr/item/UZERU_2021_55_1_a3/

[1] A. Gelman, J. B. Carlin, H. S. Stern, D. B. Dunson, A. Vehtari, D. B. Rubin, Bayesian Data Analysis. Chapman and Hall/CRC Texts in Statistical Science, 3rd ed., CRC Press, 2013, 639 pp. | MR | Zbl

[2] M. R. Hasan, A. R. Baizid, “Bayesian Estimation under Different Loss Functions Using Gamma Prior for the Case of Exponential Distribution”, Journal of Scientific Research. Section A: Physical and Mathematical Sciences, 9:1 (2017), 484–501 | DOI

[3] Th. M. Donovan, R. M. Mickey, Bayesian Statistics for Beginners: a Step-by-step Approach, 2018 | DOI | MR

[4] T. Hastie, R. Tibshirani, J. Friedman, The Elements of Statistical Learning: Data Mining, Inference and Prediction, Springer, N.Y., 2008 | DOI | MR | Zbl

[5] O. Martin, Bayesian Analysis with Python: Introduction to Statistical Modeling and Probabilistic Programming using PyMC3 and ArviZ, 2018