Tailored ensembles of neural networks optimize sensitivity to stimulus statistics

Abstract

The dynamic range of stimulus processing in living organisms is much larger than a single neural network can explain. For a generic, tunable spiking network we derive that while the dynamic range is maximal at criticality, the interval of discriminable intensities is very similar for any network tuning due to coalescence. Compensating coalescence enables adaptation of discriminable intervals. Thus, we can tailor an ensemble of networks optimized to the distribution of stimulus intensities, e.g., extending the dynamic range arbitrarily. We discuss potential applications in machine learning.

arXiv comment6 pages plus supplemental material

References in corpus (9)

• Optimal Dynamical Range of Excitable Networks at Criticality
• Colloquium: Criticality and dynamical scaling in living systems
• Predicting criticality and dynamic range in complex networks: effects of topology
• Inferring collective dynamical states from widely unobserved systems
• Inhibition causes ceaseless dynamics in networks of excitable nodes
• Homeostatic plasticity and external input shape neural network dynamics
• Dynamic Adaptive Computation: Tuning network states to task requirements
• Description of spreading dynamics by microscopic network models and macroscopic branching processes can differ due to coalescence
• Stimulus sensitivity of a spiking neural network model

Cited by in corpus (9)

• Control of criticality and computation in spiking neuromorphic networks with plasticity
• Self-organization toward criticality by synaptic plasticity
• Criticality in spreading processes without time-scale separation and the critical brain hypothesis
• Description of spreading dynamics by microscopic network models and macroscopic branching processes can differ due to coalescence
• MR. Estimator, a toolbox to determine intrinsic timescales from subsampled spiking activity
• Homeostatic Criticality in Neuronal Networks
• Sampling effects and measurement overlap can bias the inference of neuronal avalanches
• When to be critical? Performance and evolvability in different regimes of neural Ising agents
• Less is different: why sparse networks with inhibition differ from complete graphs