Geodesic
On asymptotic behavior of some class of random matrix iterations
Fundamentalʹnaâ i prikladnaâ matematika, Tome 1 (1995) no. 4, pp. 1009-1018.

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

In the paper iterations $J_{m+1}=J_m-\varepsilon J_mL_{S_m}J_m$, $m=0,1,2,\ldots$; $\varepsilon>0$ are considered. $J_m$ and $L_{S_m}$ are selfadjoint operators on $\mathbb R^N$, $L_{S_m}=(\cdot,S_m)S_m$, with $S_m$ being independent identically distributed random vectors which satisfy some additional conditions. Initial opetator $J_0$ is nonrandom. Asymptotic behavior of the rescaled operator $\tilde{J_m}=\|J_m\|^{-1}J_m$ is examined. Problems of this type appear in neural network theory when studying REM sleep phenomenon. It is proven that one of the following three relations holds almost surely: I. $\lim_{m\to\infty}\tilde{J}_m=P_{\mathcal L}$; II. $\lim_{m\to\infty}\tilde{J}_m=-P_{\xi}$; III. $J_m=0$ starting from some $m_0$; here $P_{\mathcal L}$ and $P_{\xi}$ are orthogonal projectors on random subspace $\mathcal L\subset\mathbb R^N$ and one-dimensional subspace spanned by random nonzero vector $\xi$, respectively. Denote $P_+(\varepsilon)$ and $P_-(\varepsilon)$ the probabilities of asymptotic behaviors I and II. For $J_0$ being nonzero positive semidefinite it is shown that $\lim_{\varepsilon\to+0}P_+(\varepsilon)=1$, $\lim_{\varepsilon\to+\infty}P_-(\varepsilon)=1$, but if $J_0$ has at least one negative eigenvalue, then $P_-(\varepsilon)\equiv1$. 
@article{FPM_1995_1_4_a10,
     author = {A. Yu. Plakhov},
     title = {On asymptotic behavior of some class of random matrix iterations},
     journal = {Fundamentalʹna\^a i prikladna\^a matematika},
     pages = {1009--1018},
     publisher = {mathdoc},
     volume = {1},
     number = {4},
     year = {1995},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/FPM_1995_1_4_a10/}
}
TY  - JOUR
AU  - A. Yu. Plakhov
TI  - On asymptotic behavior of some class of random matrix iterations
JO  - Fundamentalʹnaâ i prikladnaâ matematika
PY  - 1995
SP  - 1009
EP  - 1018
VL  - 1
IS  - 4
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/FPM_1995_1_4_a10/
LA  - ru
ID  - FPM_1995_1_4_a10
ER  - 
%0 Journal Article
%A A. Yu. Plakhov
%T On asymptotic behavior of some class of random matrix iterations
%J Fundamentalʹnaâ i prikladnaâ matematika
%D 1995
%P 1009-1018
%V 1
%N 4
%I mathdoc
%U http://geodesic.mathdoc.fr/item/FPM_1995_1_4_a10/
%G ru
%F FPM_1995_1_4_a10
A. Yu. Plakhov. On asymptotic behavior of some class of random matrix iterations. Fundamentalʹnaâ i prikladnaâ matematika, Tome 1 (1995) no. 4, pp. 1009-1018. http://geodesic.mathdoc.fr/item/FPM_1995_1_4_a10/

[1] J. J. Hopfield, D. I. Feinstein, R. G. Palmer, ““Unlearning” has a stabilizing effect in collective memories”, Nature, 304 (1983), 158–159 | DOI

[2] U. Wimbauer, N. Klemmer, J. L. van Hemmen, “Universality of unlearning”, Neural Networks, 7 (1994), 261 | DOI

[3] A. Yu. Plakhov, S. A. Semenov, “Neural networks: iterative unlearning algorithm converging to the projector rule matrix”, J. Phys. I France, 4 (1994), 253–260 | DOI