Transcript Design Overview - Computer Science & Engineering

NeoCortical Builder:
A Web Based Front End
for NCS
Jakub Berlinski Marlon D. Chavez, Cameron Rowe,
Nathan M. Jordan, Devyani Tanna, Roger V. Hoang,
Sergiu M. Dascalu, Laurence C. Jayet Bray, Frederick C Harris, Jr.
Brain Computation Lab
Department of Computer Science and Engineering
University of Nevada, Reno
Overview




Introduction
NCS Overview
Design Overview
Conclusion and Future Work
Human Brain


Neurons : ~ 8.6 x 10^10 (86 Billion)
Synapses: ~ 1x 10^14 (100 Trillion)
Neurons
Fun Fact - Longest axons in human body run from
base of the spinal cord to the big toe of each foot!
Channels
Signal Propagation
Levels of Organization of
Modeling
Problem


Creating a brain model is hard
Old methods are outdated or use text based input
Problem: Other Simulators
NEURON’s GUI
Problem (cont)


Current brain simulators require extraneous methods
during the setup process to be fully functional
NCS cannot easily be accessed remotely to build
models, run simulations, and view output
What is NCS?


Developed and maintained by the UNR Brain
Computation Laboratory
The NeoCortical Simulator is designed for
modeling large-scale neural networks and systems




Can model millions of neurons in real time
Open source
Runs on a heterogeneous cluster of CPUs and
NVIDIA GPUs
First simulator to support real-time neurorobotics
application
Hodgkin-Huxley Neurons
(Added in NCS 7.0)




Biologically accurate
Developed in 1952 by Alan
Hodgkin and Andrew Huxley
from their experiments on the
giant axon of a squid
Set of four differential
equations
Three variables n, m, h
Hodgkin-Huxley (cont)
Leaky Integrate-and-Fire

Comprised of






Sub-threshold leaky-integrator dynamic
Firing threshold
Reset mechanism
Leakage Channels
Drive the neuron to higher
voltage
Let the voltage decay to its
resting potential
Izhikevich





Created by Eugene M. Izhikevich
Published in 2003
Most Simplistic
Computationally efficient and captures large
variety of response properties of real neurons
Only 6 variables!
Izhikevich (Added in NCS 6.0)
Image Source:
Izhikevich Output
Solution

NeoCortical Builder (NCB)


Graphical User Interface for easy construction of brain
models and simulation parameters
Web based application for easy access from any location or
device
Design Overview
Technologies used in NCB

Core




HTML5
CSS3
JavaScript
Python

Libraries Used






JavaScript
jQuery
AngularJS
Bootstrap
X-editable
Flask
Design Overview -- Brain Builder
Design Overview -- Brain Builder
Design Overview -- Brain Builder
Design Overview -- Brain Builder
Design Overview -- Simulation
Builder
Design Overview -- Simulation
Builder
Conclusion


NCB Provides a front end interface for the
NeoCortical Simulator
Easy to use web based application with
intuitive controls
Future Work


Complete full front end application by merging
NCB with NCR and Virtual Robot
NCB



Brain Builder
Simulation Builder
NCR

Reports
 Model

Database
Virtual Robot
NeoCortical Builder: A
Web Based Front End
for NCS
Jakub Berlinski Marlon D. Chavez, Cameron Rowe,
Nathan M. Jordan, Devyani Tanna, Roger V. Hoang,
Sergiu M. Dascalu, Laurence C. Jayet Bray, Frederick C Harris, Jr.
Brain Computation Lab
Department of Computer Science and Engineering
University of Nevada, Reno