 |
Y. S. Popkov On Some Properties of Randomized Machine Learning Procedures in the Presence of Noisy Data |
 |
|
Abstract.
We study various models of measuring noises in the procedures of randomized entropy estimation of probability density functions: additive and multiplicative, measuring noises at the input and output of the object’s model. The properties of entropy-optimal probability density functions are studied, it is shown that the measurement noises corresponding to them are heteroscedastic.
Keywords:
entropy estimation, density functions, Lagrange multipliers, heteroscedastic noise, variation models.
PP. 89-95.
DOI 10.14357/20718632230209 References
1. Popkov Yu.S., Popkov A.Yu., Dubnov Yu.A. Entropy Randomization in Mashine Learning, 2023, CRC Press, Taylor & Francis Group. 2. Shannon C.E. Mathematical Theory of Communication. 1948, The Bell System Technical Journal, v.27, p.373- 423, 623-656. 3. Jaynes E.T. Information theory and statistical Mechanics. Physical Review, 1957, v.104(4), p.620-630. 4. Jaynes E.T. Gibbs vs Boltzmann entropy. American Journal of Physics, 1965, v.33, p.391-398. 5. Rosenkrantz R.D., Jaynes E.T. Paper on Probability, Statistics,and Statistical Physics. Kluwer Academic Pablishers, 1989. 6. Popkov Yu.S. Qualitative Properties of Random Maximum Entropy Estimates of Probability Density Functions. Mathematics, 2021, 9, 548, doi.org/10.3390/math9050548. 7. Bollerslev T., Engle R.F., Nelson D.B. ARCH Models. In: Engle R.F. and McFadden D.C.,eds. Handbook of Econometrics, 1994, Elsevier Science, Amsterdam, p.2961-3038. 8. Cai T.T., Wang L. Adaptive variance function estimation in heteroscedastic nonparametric regression. Ann. Stat., 2008, v. 36(5), p. 2025-2054, doi:10.1214/07 AOS509. 9. Oreshko N.I. Vosstanovlenie zakona heteroskedaticheskogo shuma pri traektornikh izmereniah na osnove veivlet-tehnologiy // Izvestia SPbLETU “LETI”, 2013, No. 9, pp. 16-21.
|
