scholarly journals Short-term plasticity in a liquid state machine biomimetic robot arm controller

Author(s):  
R. de Azambuja ◽  
F. B. Klein ◽  
S. V. Adams ◽  
M. F. Stoelen ◽  
A. Cangelosi
Author(s):  
Shiying Wang ◽  
Ziyang Kang ◽  
Lei Wang ◽  
Shiming Li ◽  
Lianhua Qu
Keyword(s):  

Author(s):  
Mohammad Z. Awad ◽  
Ryan J. Vaden ◽  
Zachary T. Irwin ◽  
Christopher L. Gonzalez ◽  
Sarah Black ◽  
...  

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Nishant Singh ◽  
Thomas Bartol ◽  
Herbert Levine ◽  
Terrence Sejnowski ◽  
Suhita Nadkarni

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Atefeh Pooryasin ◽  
Marta Maglione ◽  
Marco Schubert ◽  
Tanja Matkovic-Rachid ◽  
Sayed-mohammad Hasheminasab ◽  
...  

AbstractThe physical distance between presynaptic Ca2+ channels and the Ca2+ sensors triggering the release of neurotransmitter-containing vesicles regulates short-term plasticity (STP). While STP is highly diversified across synapse types, the computational and behavioral relevance of this diversity remains unclear. In the Drosophila brain, at nanoscale level, we can distinguish distinct coupling distances between Ca2+ channels and the (m)unc13 family priming factors, Unc13A and Unc13B. Importantly, coupling distance defines release components with distinct STP characteristics. Here, we show that while Unc13A and Unc13B both contribute to synaptic signalling, they play distinct roles in neural decoding of olfactory information at excitatory projection neuron (ePN) output synapses. Unc13A clusters closer to Ca2+ channels than Unc13B, specifically promoting fast phasic signal transfer. Reduction of Unc13A in ePNs attenuates responses to both aversive and appetitive stimuli, while reduction of Unc13B provokes a general shift towards appetitive values. Collectively, we provide direct genetic evidence that release components of distinct nanoscopic coupling distances differentially control STP to play distinct roles in neural decoding of sensory information.


Author(s):  
Mario Antoine Aoun ◽  
Mounir Boukadoum

The authors implement a Liquid State Machine composed from a pool of chaotic spiking neurons. Furthermore, a synaptic plasticity mechanism operates on the connection weights between the neurons inside the pool. A special feature of the system's classification capability is that it can learn the class of a set of time varying inputs when trained from positive examples only, thus, it is a one class classifier. To demonstrate the applicability of this novel neurocomputing architecture, the authors apply it for Online Signature Verification.


Sign in / Sign up

Export Citation Format

Share Document