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

Liquid State Machines are brain inspired spiking neural networks (SNNs) withrandom reservoir connectivity and bio-mimetic neuronal and synaptic models.Reservoir computing networks are proposed as an alternative to deep neuralnetworks to solve temporal classification problems. Previous studies suggest2nd order (double exponential) synaptic waveform to be crucial for achievinghigh accuracy for TI-46 spoken digits recognition. The proposal of long-timerange (ms) bio-mimetic synaptic waveforms is a challenge to compact and powerefficient neuromorphic hardware. In this work, we analyze the role of synapticorders namely: {\delta} (high output for single time step), 0th (rectangularwith a finite pulse width), 1st (exponential fall) and 2nd order (exponentialrise and fall) and synaptic timescales on the reservoir output response and onthe TI-46 spoken digits classification accuracy under a more comprehensiveparameter sweep. We find the optimal operating point to be correlated to anoptimal range of spiking activity in the reservoir. Further, the proposed 0thorder synapses perform at par with the biologically plausible 2nd ordersynapses. This is substantial relaxation for circuit designers as synapses arethe most abundant components in an in-memory implementation for SNNs. Thecircuit benefits for both analog and mixed-signal realizations of 0th ordersynapse are highlighted demonstrating 2-3 orders of savings in area and powerconsumptions by eliminating Op-Amps and Digital to Analog Converter circuits.This has major implications on a complete neural network implementation withfocus on peripheral limitations and algorithmic simplifications to overcomethem.

Hardware-Friendly Synaptic Orders and Timescales in Liquid State Machines for Speech Classification