BRN 2.63% 37.0¢ brainchip holdings ltd

Short-term synaptic plasticity optimally models continuous environments

  1. 2,638 Posts.
    lightbulb Created with Sketch. 1956
    https://arxiv.org/abs/2009.06808

    Short-term synaptic plasticity optimally models continuous environments

    Timoleon Moraitis, Abu Sebastian, Evangelos Eleftheriou (IBM Research - Zurich)

    Biological neural networks operate with extraordinary energy efficiency, owing to properties such as spike-based communication and synaptic plasticity driven by local activity. When emulated in silico, such properties also enable highly energy-efficient machine learning and inference systems. However, it is unclear whether these mechanisms only trade off performance for efficiency or rather they are partly responsible for the superiority of biological intelligence. Here, we first address this theoretically, proving rigorously that indeed the optimal prediction and inference of randomly but continuously transforming environments, a common natural setting, relies on adaptivity through short-term spike-timing dependent plasticity, a hallmark of biological neural networks. Secondly, we assess this theoretical optimality via simulations and also demonstrate improved artificial intelligence (AI). For the first time, a largely biologically modelled spiking neural network (SNN) surpasses state-of-the-art artificial neural networks (ANNs) in all relevant aspects, in an example task of recognizing video frames transformed by moving occlusions. The SNN recognizes the frames more accurately, even if trained on few, still, and untransformed images, with unsupervised and synaptically-local learning, binary spikes, and a single layer of neurons - all in contrast to the deep-learning-trained ANNs. These results indicate that on-line adaptivity and spike-based computation may optimize natural intelligence for natural environments. Moreover, this expands the goal of exploiting biological neuro-synaptic properties for AI, from mere efficiency, to computational supremacy altogether.

    https://hotcopper.com.au/data/attachments/2483/2483867-bc38efd94ab23294ddc31e2e6bd5294f.jpg


    don't let the pig gas light you
    Last edited by uiux: 18/09/20
 
watchlist Created with Sketch. Add BRN (ASX) to my watchlist
(20min delay)
Last
37.0¢
Change
-0.010(2.63%)
Mkt cap ! $610.0M
Open High Low Value Volume
38.0¢ 38.0¢ 36.0¢ $2.669M 7.210M

Buyers (Bids)

No. Vol. Price($)
51 1276604 36.5¢
 

Sellers (Offers)

Price($) Vol. No.
37.0¢ 231500 8
View Market Depth
Last trade - 13.57pm 27/10/2020 (20 minute delay) ?
(live)
Last
36.8¢
  Change
-0.010 ( 3.77 %)
Open High Low Volume
38.0¢ 38.0¢ 36.0¢ 5192420
Last updated 14.16pm 27/10/2020 (live) ?
BRN (ASX) Chart
arrow-down-2 Created with Sketch. arrow-down-2 Created with Sketch.