Searcharxiv⌕ Search

SEARCH · Searcharxiv

Results for “cond-mat.dis-nn”

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 109 records · Page 6Linked to original sources
arXiv2009

Organization and Complexity in a Nested Hierarchical Spin-Glass like Social Space

In this paper, we mathematically formulate the interaction and dynamics of a hierarchical complex social system where each agent is seen as an array of many variables. The formation of identities and modifications can be studied by using interaction physics in an energy landscape. The spin glass Hamiltonian is modified to suit our needs in light of complexity. The expression of variables subject to weights and topology is considered. Interactions within the same hierarchy level as well as the effect of one hierarchy level on another are studied. The effect of having many variables associated with a single agent brings about interesting dynamics. The persistence of the identity units subject to stiffness, continuous changes and also sudden reorganizations is discussed.

physics.soc-ph↗
arXiv2010

Anderson Localization of Classical Waves in Weakly Scattering Metamaterials

We study the propagation and localization of classical waves in one-dimensional disordered structures composed of alternating layers of left- and right-handed materials (mixed stacks) and compare them to the structures composed of different layers of the same material (homogeneous stacks). For weakly scattering layers, we have developed an effective analytical approach and have calculated the transmission length within a wide region of the input parameters. When both refractive index and layer thickness of a mixed stack are random, the transmission length in the long-wave range of the localized regime exhibits a quadratic power wavelength dependence with the coefficients different for mixed and homogeneous stacks. Moreover, the transmission length of a mixed stack differs from reciprocal of the Lyapunov exponent of the corresponding infinite stack. In both the ballistic regime of a mixed stack and in the near long-wave region of a homogeneous stack, the transmission length of a realization is a strongly fluctuating quantity. In the far long-wave part of the ballistic region, the homogeneous stack becomes effectively uniform and the transmission length fluctuations are weaker. The crossover region from the localization to the ballistic regime is relatively narrow for both mixed and homogeneous stacks. In mixed stacks with only refractive-index disorder, Anderson localization at long wavelengths is substantially suppressed, with the localization length growing with the wavelength much faster than for homogeneous stacks. The crossover region becomes essentially wider and transmission resonances appear only in much longer stacks. All theoretical predictions are in an excellent agreement with the results of numerical simulations.

cond-mat.dis-nn↗
arXiv2010

Competitive cluster growth on networks: complex dynamics and survival strategies

We extend the study of a model of competitive cluster growth in an active medium to a basis of networks; this is done by adding nonlocal connections with probability $p$ to sites on a regular lattice, thus enabling one to interpolate between regularity and full randomness. The model on networks demonstrates high sensitivity to small changes in initial configurations, which we characterize using damage spreading. The main focus of this paper is, however, the devising of survival strategies through selective networking, to alter the the fate of an arbitrarily chosen cluster: whether this be to revive a dying cluster to life, or to make a weak survivor into a stronger one. Although such goals are typically achieved by networking with relatively small clusters, our results suggest that it ought to be possible also to network with peers and larger clusters. The main indication of this comes from the probability distributions of mass differences between survivors and their immediate neighbours, which show an astonishing universality; they suggest strategies for winning against the odds.

cond-mat.stat-mech↗
arXiv2010

Interactions between Intrinsic and Stimulus-Evoked Activity in Recurrent Neural Networks

Trial-to-trial variability is an essential feature of neural responses, but its source is a subject of active debate. Response variability (Mast and Victor, 1991; Arieli et al., 1995 & 1996; Anderson et al., 2000 & 2001; Kenet et al., 2003; Petersen et al., 2003a & b; Fiser, Chiu and Weliky, 2004; MacLean et al., 2005; Yuste et al., 2005; Vincent et al., 2007) is often treated as random noise, generated either by other brain areas, or by stochastic processes within the circuitry being studied. We call such sources of variability external to stress the independence of this form of noise from activity driven by the stimulus. Variability can also be generated internally by the same network dynamics that generates responses to a stimulus. How can we distinguish between external and internal sources of response variability? Here we show that internal sources of variability interact nonlinearly with stimulus-induced activity, and this interaction yields a suppression of noise in the evoked state. This provides a theoretical basis and potential mechanism for the experimental observation that, in many brain areas, stimuli cause significant suppression of neuronal variability (Werner and Mountcastle, 1963; Fortier, Smith and Kalaska, 1993; Anderson et al., 2000; Friedrich and Laurent, 2004; Churchland et al., 2006; Finn, Priebe and Ferster, 2007; Mitchell, Sundberg and Reynolds, 2007; Churchland et al., 2009). The combined theoretical and experimental results suggest that internally generated activity is a significant contributor to response variability in neural circuits.

q-bio.NC↗
arXiv2011

Percolation Thresholds of the Fortuin-Kasteleyn Cluster for a Potts Gauge Glass Model on Complex Networks: Analytical Results on the Nishimori Line

It was pointed out by de Arcangelis et al. [Europhys. Lett. 14 (1991), 515] that the correct understanding of the percolation phenomenon of the Fortuin-Kasteleyn cluster in the Edwards-Anderson model is important since a dynamical transition, which is characterized by a parameter called the Hamming distance or damage, and the percolation transition are related to a transition for a signal propagating between spins. We show analytically the percolation thresholds of the Fortuin-Kasteleyn cluster for a Potts gauge glass model, which is an extended model of the Edwards-Anderson model, on random graphs with arbitary degree distributions. The results are shown on the Nishimori line. We also show the results for the infinite-range model.

cond-mat.dis-nn↗
arXiv2010

A Monte Carlo Algorithm for Universally Optimal Bayesian Sequence Prediction and Planning

The aim of this work is to address the question of whether we can in principle design rational decision-making agents or artificial intelligences embedded in computable physics such that their decisions are optimal in reasonable mathematical senses. Recent developments in rare event probability estimation, recursive bayesian inference, neural networks, and probabilistic planning are sufficient to explicitly approximate reinforcement learners of the AIXI style with non-trivial model classes (here, the class of resource-bounded Turing machines). Consideration of the effects of resource limitations in a concrete implementation leads to insights about possible architectures for learning systems using optimal decision makers as components.

nlin.AO↗
arXiv2010

Mean Field Theory of the Three-Dimensional Dipole Superspin Glasses

We study the three-dimensional system of magnetic nanoparticle dipoles randomly oriented along quenched easy axes. Directions of the magnetic momenta are described by the Ising variables which allow the momenta to flip along their random orientations. Using the standard mean-field approximation and the replica technique it is shown that the system undergoes a finite temperature phase transition into a spin-glass phase.

cond-mat.dis-nn↗
arXiv2010

Statistical Mechanical Analysis of a Typical Reconstruction Limit of Compressed Sensing

We use the replica method of statistical mechanics to examine a typical performance of correctly reconstructing $N$-dimensional sparse vector $bx=(x_i)$ from its linear transformation $by=bF bx$ of $P$ dimensions on the basis of minimization of the $L_p$-norm $||bx||_p= lim_{epsilon to +0} sum_{i=1}^N |x_i|^{p+epsilon}$. We characterize the reconstruction performance by the critical relation of the successful reconstruction between the ratio $alpha=P/N$ and the density $rho$ of non-zero elements in $bx$ in the limit $P,,N to infty$ while keeping $alpha sim O(1)$ and allowing asymptotically negligible reconstruction errors. We show that the critical relation $alpha_c(rho)$ holds universally as long as $bF^{rm T}bF$ can be characterized asymptotically by a rotationally invariant random matrix ensemble and $bF bF^{rm T}$ is typically of full rank. This supports the universality of the critical relation observed by Donoho and Tanner ({em Phil. Trans. R. Soc. A}, vol.~367, pp.~4273--4293, 2009; arXiv: 0807.3590) for various ensembles of compression matrices.

cs.IT↗
arXiv2010

Statistical Mechanical Analysis of Compressed Sensing Utilizing Correlated Compression Matrix

We investigate a reconstruction limit of compressed sensing for a reconstruction scheme based on the L1-norm minimization utilizing a correlated compression matrix with a statistical mechanics method. We focus on the compression matrix modeled as the Kronecker-type random matrix studied in research on multi-input multi-output wireless communication systems. We found that strong one-dimensional correlations between expansion bases of original information slightly degrade reconstruction performance.

cs.IT↗
arXiv2010

Interpreting Quantum Chromodynamics from Spin Glass- and Polymer Analogies

In an informal way some kind of Ising Lattice QCD is introduced which allows to interprete and discuss the well-known theory of quantum chromodynamics (confinement, quarks and gluons, etc.) from simple phenomena of magnetism and polymer physics. Also the viewpoint of duality is stressed. Moreover, the non-abelian character of the Lie group SU(3) is important. The essential arguments for confinement and asymptotic freedom are given at the very end of the manuscript, after an appendix.

cond-mat.dis-nn↗
arXiv2010

Statistical model of charge transport in colloidal quantum dot array

A new statistical model of charge transport in colloidal quantum dot arrays is proposed. It takes into account Coulomb blockade forbidding multiple occupancy of nanocrystals and influence of energetic disorder of interdot space. The model explains power law current transients and the presence of memory effect. The fractional differential analogue of the Ohm law is found phenomenologically for nanocrystal arrays. The model combines ideas that were considered as conflicting by other authors: the Scher-Montroll idea about power law distribution of waiting times in localized states for disordered semiconductors is applied with taking into account Coulomb blockade, Novikov's condition about asymptotical power law distribution of time intervals between successful current pulses in conduction channels is fulfilled, carrier injection blocking predicted by Ginger and Greenham takes place.

cond-mat.dis-nn↗
arXiv2010

Phase fluctuations in finite thickness disordered superconducting thin films

Phase fluctuations in finite thickness layered superconducting films are studied theoretically. The model consists of a set of layers, coupled to each other via a gradient-like term in the phase-only action. It is shown that the effective phase stiffness of the whole system is proportional to its thickness (i.e. the number of individual layers). This result leads to a superconductor-insulator transition and an inverse dependence of $T_c$ on thickness,$T_c\sim d^{-1}$, both in accord with the experimental findings. The relevance to disordered thin films is discussed.

cond-mat.dis-nn↗
arXiv2010

More than two equally probable variants of signal in Kauffman networks as an important overlooked case, negative feedbacks allow life in chaos

There are three main aims of this paper. 1- I explain reasons why I await life to lie significantly deeper in chaos than Kauffman approach does, however still in boundary area near `the edge of chaos and order'. The role of negative feedbacks in stability of living objects is main of those reasons. In Kauffman's approach regulation using negative feedbacks is not considered sufficiently, e.g. in gene regulatory model based on Boolean networks, which indicates therefore not proper source of stability. Large damage avalanche is available only in chaotic phase. It models death in all living objects necessary for Darwinian elimination. It is the first step of my approach leading to structural tendencies which are effects of adaptive evolution of dynamic complex (maturely chaotic) networks. 2- Introduction of s>=2 equally probable variants of signal (state of node in Kauffman network) as interpretively based new statistical mechanism (RSN) instead of the bias p - probability of one of signal variants used in RBN family and RNS. It is also different than RWN model. For this mechanism which can be treated as very frequent, ordered phase occurs only in exceptional cases but for this approach the chaotic phase is investigated. Annealed approximation expectations and simulations of damage spreading for different network types (similar to CRBN, FSRBN and EFRBN but with s>=2) are described. Degree of order in chaotic phase in dependency of network parameters and type is discussed. By using such order life evolve. 3- A simplified algorithm called `reversed-annealed' for statistical simulation of damage spreading is described. It is used for simulations presented in this and next papers describing my approach.

cond-mat.dis-nn↗
arXiv2010

Experimental observation of speckle instability in Kerr random media

In a disordered medium with Kerr-type nonlinearity, the transmitted speckle pattern was predicted to become unstable, as a result of the positive feedback between the intensity fluctuations and the nonlinear dependence of the local refractive index. We present the first experimental evidence of speckle instability of light transversally scattered in a liquid crystal cell. A two-dimensional controlled disorder is imprinted in the medium through a suitable illumination of a photoconductive wall of the cell, whereas the nonlinear response is obtained through optical reorientation of the liquid crystal molecules. Speckle patterns with oscillating intensity are observed above a critical threshold, which is found to depend on the scattering mean free path, as predicted by theory.

cond-mat.dis-nn↗
arXiv2010

Phase Transitions in Disordered Systems

We face the problem of phase transitions in diluted systems both from theoretical and numerical sides. We study the effects of quenched site-dilution in classical models (Heisenberg, Ising and Potts) in 2, 3, and 4 dimensions both by using the Renormalization Group and numerical simulations in the canonical and microcanonical ensembles. We propose and check a new formulation of the Finite Size Scaling ansatz (FSS) inside the microcanonical ensemble. We use microcanonical simulations to obtain a clear fist-order behavior for the diluted Potts model in 3D, estimating the tricritical dilution. We perform large-scale simulations of the 3D diluted Heisenberg model, checking its self-averaging properties. Finally we study the 4D diluted Ising model obtaining from the FSS of the specific heat a clear differentiation between the existing conflicting theories. We also compiled a large number of appendix that we expect to be used as future reference.

cond-mat.dis-nn↗