The equivalence condition for two important adaptation algorithms and its relation to efficient estimates of probability distribution parameters
A comparison is made between the convergence rates of two important algorithms in adaptation theory; one of them minimizes the empirical risk, while the other minimizes the Bayes a posteriori risk and is calculated according to the stipulated system of probability distributions which depend on a set of unknown parameters. The class of distributions is established for which the convergence rates of the indicated algorithms coincide. The relation between the conditions governing the indicated coincidence and the conditions governing the existence of efficient estimates of the unknown probability distribution parameters is investigated.
KeywordsProbability Distribution Convergence Rate Unknown Parameter Equivalence Condition Quantum Electronics
Unable to display preview. Download preview PDF.
- 1.Ya. Z. Tsypkin, Adaptation and Learning in Automatic Systems [in Russian], Nauka, Moscow (1968).Google Scholar
- 2.R. L. Stratonovich, Avtomat. i Telemekh., No. 1, 96 (1968).Google Scholar
- 3.R. L. Stratonovich, Avtomat. i Telemekh., No. 5, 97 (1970).Google Scholar
- 4.J. Neumann, Matematika,8, No. 2, 113 (1964).Google Scholar
- 5.R. L. Stratonovich, Izv. Akad. Nauk SSSR, Tekh. Kibernetika, No. 6 (1969).Google Scholar
- 6.S. Rao, Linear Statistical Methods and Their Application [in Russian], Nauka, Moscow (1968).Google Scholar
- 7.G. Robbins, Matematika,8, No. 2, 133 (1964).Google Scholar