Searcharxiv⌕ Search

arXiv · cond-mat/0202190

Threshold Disorder as a Source of Diverse and Complex Behavior in Random Nets

Abstract

We study the diversity of complex spatio-temporal patterns in the behavior of random synchronous asymmetric neural networks (RSANNs). Special attention is given to the impact of disordered threshold values on limit-cycle diversity and limit-cycle complexity in RSANNs which have `normal' thresholds by default. Surprisingly, RSANNs exhibit only a small repertoire of rather complex limit-cycle patterns when all parameters are fixed. This repertoire of complex patterns is also rather stable with respect to small parameter changes. These two unexpected results may generalize to the study of other complex systems. In order to reach beyond this seemingly-disabling `stable and small' aspect of the limit-cycle repertoire of RSANNs, we have found that if an RSANN has threshold disorder above a critical level, then there is a rapid increase of the size of the repertoire of patterns. The repertoire size initially follows a power-law function of the magnitude of the threshold disorder. As the disorder increases further, the limit-cycle patterns themselves become simpler until at a second critical level most of the limit cycles become simple fixed points. Nonetheless, for moderate changes in the threshold parameters, RSANNs are found to display specific features of behavior desired for rapidly-responding processing systems: accessibility to a large set of complex patterns.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Patrick C. McGuire, Henrik Bohr, John W. Clark, Robert Haschke, Chris Pershing, Johann Rafelski. 2002-02-12. Threshold Disorder as a Source of Diverse and Complex Behavior in Random Nets. https://arxiv.org/abs/cond-mat/0202190

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

KEEP EXPLORING

Related papers

Embedding stochastic dynamics in latent manifolds of recurrent neural networks

Recurrent neural networks (RNNs) provide a theoretical framework for understanding computation in biological neural circuits through their ability to support rich time-dependent behaviour. Whilst high-dimensional by construction, they can be used to encode low-dimensional dynamical systems, for example by restricting to low-rank connectivity, hence mirroring the emergence of latent neural manifolds found in empirical neural data. Here, we exploit a relationship between RNNs and a class of universal approximators to train networks that can represent arbitrary, nonlinear stochastic differential equations (SDEs), with given drift and diffusion coefficients, within a low-dimensional latent subspace. We show that RNNs can faithfully embed the drift and diffusion of a given SDE, including nonlinear and nonequilibrium dynamics such as chaotic attractors. We apply this approach by constructing RNN realisations of stochastic systems that transiently explore various attractors through both input-driven switching and autonomous transitions driven by nonequilibrium currents, which we interpret as models of associative and sequential (episodic) memory. To elucidate how these dynamics are encoded in the network, we introduce decompositions of the RNN based on its asymmetric connectivity and its time-irreversibility. Our results show that neural populations can implement a broad class of stochastic dynamics within low-dimensional manifolds, unifying ideas from the dynamics of neural computation, the theory of neural manifolds, associative memory, and nonequilibrium statistical mechanics.

cond-mat.dis-nn↗

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↗

Cup and Cap Topological Neural Network

Topological Deep Learning (TDL) is designed to learn features associated not only with nodes but also with edges, triangles and higher dimensional simplices. However, most TDL approaches, being based on boundary operators and Hodge Laplacians, have the limitation that features and signals cannot be lifted or lowered across more than one dimension per layer. To overcome this limitation, in this work, we propose the adoption of the cup and cap products. Specifically, we formulate the Cup and Cap Topological Neural Network (CCNN), a topological deep learning architecture designed to learn node-based variables (or 0-cochains) by taking into account their many-body interactions (e.g. triangles) present in the data. We validate CCNN by assessing its performance on the TopoBench datasets, demonstrating its competitive performance with respect to other simplicial complex neural networks.

cond-mat.dis-nn↗