02-08-2011, 02:32 PM
Presented by:
JISHNU G KANNAN
[attachment=14991]
THE ARTIFICIAL BRAIN IN REALITY:
INTRODUCTION
First comprehensive attempt to reverse-engineer the mammalian brain.
We lose the knowledge of a brain when the body is destroyed after death .
What would happen if we create an artificial brain and upload the contents of natural brain into it??
WHAT IS BLUE BRAIN?
Scientists are in research to create an artificial brain that can think, respond and take decisions.
A network of artificial nerves is evolving right now in a Swiss super computer.
Blue brain will offer a better understanding of human consciousness.
Focused on biomedical applications.
REAL BRAIN BEHIND THE VIRTUAL BRAIN
HENRY MARKRAM
Project director of The Blue Brain Project
MILESTONES
In July ’05, EPFL and IBM announced a new research initiative- a project to create a functional model of brain.
In ’06, the project created a model of the basic functional unit of the brain- The NeoCorticalColumn(NCC).
Initial goal of project was completed in dec ‘06- Simulation of rat’s NCC.
In Nov ‘07, end of Phase-I was reported.
COMBINING COMPUTER SCIENCE AND NEURO SCIENCE
The Brain:
Cerebral Cortex
Convoluted grey matter that makes up 80% of the human brain.
Responsible for our ability to think, respond, communicate and plan for the future.
BIOLOGICAL NEURON
Biological neuron: Basic component of brain.
Connected by dendrites and axons
Brain process information by sending electrical signals from neuron to neuron.
NEOCORTICAL COLUMN(NCC)
In cortex, neurons are organized into basic functional units each containing about 10000 neurons.
Operates much like microcircuits in a computer.
ARTIFICIAL NEURONS
2 modes of operation
In training mode, the neuron can be trained to fire (or not).
In using mode, when a taught input pattern is detected at the input, its associated output becomes the current output.
THREE ASPECTS OF COMPUTATION
Neural computing must achieve balance between:
Processing functions
Communicating functions
Storage functions
These are the three aspects of computation
BUILDING THE MICRO CIRCUITRY
Modeling neurons and connections: Blue Gene recreates each neurons in NCC and fixes the synapse locations.
Modeling column: Result of calculations is a re-creation of the NCC.
SIMULATION OF MICRO-CIRCUITRY
8192 processors of the Blue Gene are pressed into service.
Parallel computation solves the complex mathematical equations that govern the electrical activity in each neuron.
Results are communicated via inter-processor communication.
Time required to simulate is twice the magnitude than the actual biological time.
BLUE GENE SUPERCOMPUTER
IBM’s Super Computer
360 TFLOPS of computing power from 8192 PowerPC CPUs each running at 700MHz and arranged in a toroidal mesh.
BLUE GENE
Built of large number of nodes, each having relatively modest clock rate.
Each node consists of a single ASIC and 2 GB memory.
Each processor can perform 4 floating point operations per cycle. Each node contains a second processor for handling message passing.
Scalable system software that supports efficient executing.
INTREPRETING THE RESULTS
Analysis of individual neurons must be repeated 1000s of times.
Visual exploration of circuit is important to analysis.
Mapping simulation data onto the morphology is for immediate verification.
NCC has 10,000 neurons.
When Impulse travels, neurons light up and change colour as they become electrically active.
Calibration is the second stage of the Blue Brain project.
ELEMENTARY BUILDING BLOCKS OF NEURAL MICROCIRCUITRY
MODELLING OF THE BRAIN
BUILDING THE BLUE COLUMN
Parse each 3D morphology and correct errors.
Statistics are used to clone neurons in each class.
Insert ion channel models to produce the array of electrical types.
EXPERIMENTAL EVIDENCE
INITIAL OBSERVATIONS
Willful and unpredictable from day one.
When first fed with electrical impulses, strange patterns with lightning like flashes appeared.
Neurons started interacting with one another.
Artificial cells used are random and unique.
ADVANTAGES
1. Remembering things without any effort.
2. Making decisions without the presence of a person.
3. Using intelligence of a person after his death. .
4. Understanding the activities of animals.
5. Allowing the deaf to hear via direct nerve
stimulation.
DISADVANTAGES
1. We become dependent upon the computer.
2. Others may use technical knowledge against us.
3. Another fear is found today with respect to human cloning.
4. Computer viruses can pose an increasingly critical threat.
APPLICATIONS
1. Gathering and testing 100 years of data
2. Understanding neocortical information processing
3. A novel tool for drug discovery for brain disorders
4. A foundation for whole brain simulations
FUTURE WORKS
Phase 1 provide the basis for increasing the resolution of the models and expanding the size of the models.
Information from the molecular and genetic level will be added to the algorithms.
Insights into how human beings think and remember.
Shed light on psychiatric disorders.
If we crack open the secret of how and why the brain does it, could lead to new breed of supercomputers.
CONCLUSION
We will be able to transfer ourselves into computers at some point.
Answers to mental retardation and brain abnormalities can be solved at earlier stages.
Diseases like Alzheimer’s and Parkinson’s that rid a person of his dignity could be eradicated totally.
JISHNU G KANNAN
[attachment=14991]
THE ARTIFICIAL BRAIN IN REALITY:
INTRODUCTION
First comprehensive attempt to reverse-engineer the mammalian brain.
We lose the knowledge of a brain when the body is destroyed after death .
What would happen if we create an artificial brain and upload the contents of natural brain into it??
WHAT IS BLUE BRAIN?
Scientists are in research to create an artificial brain that can think, respond and take decisions.
A network of artificial nerves is evolving right now in a Swiss super computer.
Blue brain will offer a better understanding of human consciousness.
Focused on biomedical applications.
REAL BRAIN BEHIND THE VIRTUAL BRAIN
HENRY MARKRAM
Project director of The Blue Brain Project
MILESTONES
In July ’05, EPFL and IBM announced a new research initiative- a project to create a functional model of brain.
In ’06, the project created a model of the basic functional unit of the brain- The NeoCorticalColumn(NCC).
Initial goal of project was completed in dec ‘06- Simulation of rat’s NCC.
In Nov ‘07, end of Phase-I was reported.
COMBINING COMPUTER SCIENCE AND NEURO SCIENCE
The Brain:
Cerebral Cortex
Convoluted grey matter that makes up 80% of the human brain.
Responsible for our ability to think, respond, communicate and plan for the future.
BIOLOGICAL NEURON
Biological neuron: Basic component of brain.
Connected by dendrites and axons
Brain process information by sending electrical signals from neuron to neuron.
NEOCORTICAL COLUMN(NCC)
In cortex, neurons are organized into basic functional units each containing about 10000 neurons.
Operates much like microcircuits in a computer.
ARTIFICIAL NEURONS
2 modes of operation
In training mode, the neuron can be trained to fire (or not).
In using mode, when a taught input pattern is detected at the input, its associated output becomes the current output.
THREE ASPECTS OF COMPUTATION
Neural computing must achieve balance between:
Processing functions
Communicating functions
Storage functions
These are the three aspects of computation
BUILDING THE MICRO CIRCUITRY
Modeling neurons and connections: Blue Gene recreates each neurons in NCC and fixes the synapse locations.
Modeling column: Result of calculations is a re-creation of the NCC.
SIMULATION OF MICRO-CIRCUITRY
8192 processors of the Blue Gene are pressed into service.
Parallel computation solves the complex mathematical equations that govern the electrical activity in each neuron.
Results are communicated via inter-processor communication.
Time required to simulate is twice the magnitude than the actual biological time.
BLUE GENE SUPERCOMPUTER
IBM’s Super Computer
360 TFLOPS of computing power from 8192 PowerPC CPUs each running at 700MHz and arranged in a toroidal mesh.
BLUE GENE
Built of large number of nodes, each having relatively modest clock rate.
Each node consists of a single ASIC and 2 GB memory.
Each processor can perform 4 floating point operations per cycle. Each node contains a second processor for handling message passing.
Scalable system software that supports efficient executing.
INTREPRETING THE RESULTS
Analysis of individual neurons must be repeated 1000s of times.
Visual exploration of circuit is important to analysis.
Mapping simulation data onto the morphology is for immediate verification.
NCC has 10,000 neurons.
When Impulse travels, neurons light up and change colour as they become electrically active.
Calibration is the second stage of the Blue Brain project.
ELEMENTARY BUILDING BLOCKS OF NEURAL MICROCIRCUITRY
MODELLING OF THE BRAIN
BUILDING THE BLUE COLUMN
Parse each 3D morphology and correct errors.
Statistics are used to clone neurons in each class.
Insert ion channel models to produce the array of electrical types.
EXPERIMENTAL EVIDENCE
INITIAL OBSERVATIONS
Willful and unpredictable from day one.
When first fed with electrical impulses, strange patterns with lightning like flashes appeared.
Neurons started interacting with one another.
Artificial cells used are random and unique.
ADVANTAGES
1. Remembering things without any effort.
2. Making decisions without the presence of a person.
3. Using intelligence of a person after his death. .
4. Understanding the activities of animals.
5. Allowing the deaf to hear via direct nerve
stimulation.
DISADVANTAGES
1. We become dependent upon the computer.
2. Others may use technical knowledge against us.
3. Another fear is found today with respect to human cloning.
4. Computer viruses can pose an increasingly critical threat.
APPLICATIONS
1. Gathering and testing 100 years of data
2. Understanding neocortical information processing
3. A novel tool for drug discovery for brain disorders
4. A foundation for whole brain simulations
FUTURE WORKS
Phase 1 provide the basis for increasing the resolution of the models and expanding the size of the models.
Information from the molecular and genetic level will be added to the algorithms.
Insights into how human beings think and remember.
Shed light on psychiatric disorders.
If we crack open the secret of how and why the brain does it, could lead to new breed of supercomputers.
CONCLUSION
We will be able to transfer ourselves into computers at some point.
Answers to mental retardation and brain abnormalities can be solved at earlier stages.
Diseases like Alzheimer’s and Parkinson’s that rid a person of his dignity could be eradicated totally.
