English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recurrent networks are a special class of artificial neural systems that usetheir internal states to perform computing tasks for machine learning. One ofits state-of-the-art developments, i.e. reservoir computing (RC), uses theinternal structure -- usually a static network with random structure -- to mapan input signal into a nonlinear dynamical system defined in a higherdimensional space. Reservoirs are characterized by nonlinear interactions amongtheir units and their ability to store information through recurrent loops,allowing to train artificial systems to learn task-specific dynamics. However,it is fundamentally unknown how the random topology of the reservoir affectsthe learning performance. Here, we fill this gap by considering a battery ofsynthetic networks -- characterized by different topological features -- and 45empirical connectomes -- sampled from brain regions of organisms belonging to 8different species -- to build the reservoir and testing the learningperformance against a prediction task with a variety of complex input signals.We find nontrivial correlations between RC performances and both the number ofnodes and rank of the covariance matrix of activation states, with performancedepending on the nature -- stochastic or deterministic -- of input signals.Remarkably, the modularity and the link density of the reservoir are found toaffect RC performances: these results cannot be predicted by models onlyaccounting for simple topological features of the reservoir. Overall, ourfindings highlight that the complex topological features characterizingbiophysical computing systems such as connectomes can be used to designefficient bio-inspired artificial neural networks.

Complex topological features of reservoirs shape learning performances in bio-inspired recurrent neural networks