Searcharxiv⌕ Search

arXiv · cond-mat/0408642

Transmission distribution, P(\ln T), of 1D disordered chain: low-T tail

Abstract

We demonstrate that the tail of transmission distribution through 1D disordered Anderson chain is a strong function of the correlation radius of the random potential, $a$, even when this radius is much shorter than the de Broglie wavelength, $k_F^{-1}$. The reason is that the correlation radius defines the phase volume of the trapping configurations of the random potential, which are responsible for the low-$T$ tail. To see this, we perform the averaging over the low-$T$ disorder configurations by first introducing a finite lattice spacing $\sim a$, and then demonstrating that the prefactor in the corresponding functional integral is exponentially small and depends on $a$ even as $a \to 0$. Moreover, we demonstrate that this restriction of the phase volume leads to the dramatic change in the shape of the tail of ${\cal P}(\ln T)$ from universal Gaussian in $\ln T $ to a simple exponential (in $\ln T $) with exponent depending on $a$. Severity of the phase-volume restriction affects the shape of the low-$T$ disorder configurations transforming them from almost periodic (Bragg mirrors) to periodically-sign-alternating (loose mirrors).

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

V. M. Apalkov, M. E. Raikh. 2004-08-29. Transmission distribution, P(\ln T), of 1D disordered chain: low-T tail. https://arxiv.org/abs/cond-mat/0408642

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Deep-layered machines have a built-in Occam's razor

Many complex systems map vast numbers of microscopic descriptions onto far fewer macroscopic outcomes. Surprisingly, these input-output maps strongly favor simple outputs. In an exactly solvable deep-layered machine, we prove that simplicity bias strengthens with every layer. The probability of an output function $F$ approaches $\Pr(F)\propto 2^{-K(F)}$, where $K$ is the Kolmogorov complexity. Our results show that hierarchical composition alone can generate an intrinsic Occam's razor, suggesting a general mechanism for simplicity bias in learning and evolution.

cond-mat.dis-nn↗

Neural Networks with Exciton-Polariton Waveguides

We propose and numerically study a scheme for neural networks using exciton-polariton waveguide meshes. These enable arbitrary programmable interlayer couplings suited to a restricted Boltzmann machine architecture where polariton phases act as stochastic network variables. We show that such a network can be trained using unsupervised Hebbian and anti-Hebbian updates. This allows the implementation of an associative memory, where the network recovers learned patterns from imperfect inputs.

cond-mat.dis-nn↗

Finite-width shift of the edge of chaos in deep neural networks

Deep neural networks train best when they are initialised at the edge of chaos, the critical line in the plane of weight and bias variances on which the distance between two inputs neither grows nor shrinks with depth. The line is known for infinitely wide networks; at finite width it moves, and how it moves has been disputed. We treat the logarithm of the distance between two inputs through a network of width $N$ as a random walk and compute its drift and variance to order $1/N$. The edge of chaos moves to a larger weight variance, by $Δσ_w^2=-a_1/(N\,\partial_{σ_w^2}\lnχ_1)$, about $8/N$ for $\tanh$ networks at a bias variance of $0.05$, confirmed by simulations of the exact finite-width recursion. The shift describes a typical network; the average over many networks is dominated by rare networks in which the distance grows, and its critical point moves the other way, which explains why calculations and simulations have disagreed. Backpropagation follows the same random walk, so the shift applies to gradients unchanged. Along the edge of chaos, the information about the similarity of two inputs is constant until the compression over the depth exceeds about one nat, which fixes where to sit on the line. The results give an initialisation rule, Crinit, that places a network on its finite-width edge of chaos without a search. On MNIST and Fashion-MNIST at width $64$ it is more accurate than the standard rules at their usual settings, by a small margin in shallow networks and a large one in deep networks, where most standard rules fail; it trains in fewer steps, and it matches or beats the standard rules with a tuned gain.

cond-mat.dis-nn↗