05-09-BrainBuilding-review

8/7/2019 05-09-BrainBuilding-review

1/30

LargeLarge--scale projects to buildscale projects to build

artificial brains: review.artificial brains: review.

WodzisawWodzisaw DuchDuch (Google:(Google: DuchDuch))

Department of Informatics,Department of Informatics,NicolausNicolaus Copernicus University,Copernicus University,

Torun, PolandTorun, Poland

School of Computer Engineering,School of Computer Engineering,NanyangNanyang TechnologicalTechnological University (NTU),University (NTU),

SingaporeSingapore

Building Artificial Brain workshop after ICANN 2005, Sept 15, 2005


2/30

Plan Motivation: are we ready for brain simulation?Motivation: are we ready for brain simulation?

Some failed attempts.Some failed attempts. Special hardware?Special hardware?

Nomad/Darwin robots, Gerald EdelmanNomad/Darwin robots, Gerald Edelman

Blue BrainBlue Brain Henry Markram, Lausanne/IBMHenry Markram, Lausanne/IBM

CCortex, Artificial Development.CCortex, Artificial Development. The Ersatz Brain Project, James AndersonThe Ersatz Brain Project, James Anderson

AiAi developing brains?developing brains?

Conscious machines: Pentti Haikonen (Nokia) & others.Conscious machines: Pentti Haikonen (Nokia) & others.

Bayesian confidence propagating network: LansnerBayesian confidence propagating network: Lansner

Artificial Mind SystemArtificial Mind System Testuya HoyaTestuya Hoya NTU projects in artificial mindsNTU projects in artificial minds

Related EU projects and initiativesRelated EU projects and initiatives

Related: consciousness is not that hard; how to get mind out ofRelated: consciousness is not that hard; how to get mind out ofbrain?brain?


3/30

Motivation: developments in computingMotivation: developments in computing

Naive estimation of the brain power:Naive estimation of the brain power:BP = 100 Hz x 10BP = 100 Hz x 101414 synapses = 10synapses = 101616 binop/s.binop/s.

Power for abstract thinking is probably much lower.Power for abstract thinking is probably much lower.

Kasparov lost in 1997 with Deep Blue machine that searched 200MKasparov lost in 1997 with Deep Blue machine that searched 200Mnodes/sec, less than 10nodes/sec, less than 101212 binop/s, on 32binop/s, on 32--processor IBM SP + 512processor IBM SP + 512specialized chess processors. This gives about 0.01% of BP.specialized chess processors. This gives about 0.01% of BP.

Kramnik (2002) reached a draw with 8Kramnik (2002) reached a draw with 8--processor Windows XP machineprocessor Windows XP machinerunning commercial version of Deep Fritz program.running commercial version of Deep Fritz program.

Supercomputer speeds have just reached > 100 Tflops, or a fewSupercomputer speeds have just reached > 100 Tflops, or a fewPetaops/sec, comparable with brain power, Grid computing arrived, butPetaops/sec, comparable with brain power, Grid computing arrived, butcomputers are far from brains complexity and processing style.computers are far from brains complexity and processing style.

In the near future 1000$ PC will have brain power.In the near future 1000$ PC will have brain power.


4/30

CComputingomputing/inteligence/inteligence


5/30

Computing costsComputing costs


6/30

Motivation: neuroscienceMotivation: neuroscience

From the Blue Brain project:From the Blue Brain project:

Scientists have been accumulating knowledge on the structure andScientists have been accumulating knowledge on the structure andfunction of the brain for the past 100 years. It is now time to startfunction of the brain for the past 100 years. It is now time to startgathering this data together in a unified model and putting it to the testgathering this data together in a unified model and putting it to the testin simulations. We still need to learn a lot about the brain before wein simulations. We still need to learn a lot about the brain before we

understand it's inner workings, but building this model should helpunderstand it's inner workings, but building this model should helporganize and accelerate this quest.organize and accelerate this quest.

The data obtained on the microstructure and function of the NCC hasThe data obtained on the microstructure and function of the NCC hasnow reached a critical level of detail that makes it possible to begin anow reached a critical level of detail that makes it possible to begin asystematic reconstruction of the NCC. The numbers and types ofsystematic reconstruction of the NCC. The numbers and types ofneurons have basically been defined, who connects to whom and howneurons have basically been defined, who connects to whom and howoften, has been worked out, and the way that most of the neuronsoften, has been worked out, and the way that most of the neuronsfunction as well as the way that the neurons communicate and learnfunction as well as the way that the neurons communicate and learnhas been extensively studied.has been extensively studied.

We therefore now have a near complete digital description of theWe therefore now have a near complete digital description of thestructural and functional rules of the NCC.structural and functional rules of the NCC.


7/30

Scheme of the brain ...Scheme of the brain ...

HighHigh--level sketch of the brain structures, with connections based onlevel sketch of the brain structures, with connections based ondifferent types of neurotransmiters marked in different colors.different types of neurotransmiters marked in different colors.


8/30

Motivation: more scienceMotivation: more science

Engineering: to be sure that we understand complex systemEngineering: to be sure that we understand complex systemwe need to build and test them.we need to build and test them.

Understanding emergent properties of neural systems: howUnderstanding emergent properties of neural systems: howhighhigh--level cognition arises from lowlevel cognition arises from low--level interactions betweenlevel interactions betweenneurons.neurons.

Removing all but a few areas of the brain will to lead toRemoving all but a few areas of the brain will to lead tofunctional system, therefore even crude simulation thatfunctional system, therefore even crude simulation thatincludes all major areas can teach us something.includes all major areas can teach us something.

Build powerful research tool for brain sciences.Build powerful research tool for brain sciences.

So far the only architecture of cognition is SOAR, based onSo far the only architecture of cognition is SOAR, based onthe idea of physical symbol processing system, originated bythe idea of physical symbol processing system, originated byNewell, Simon & developed over the last 25 years. SOAR andNewell, Simon & developed over the last 25 years. SOAR andACTACT--R were very successful in explaining different features ofR were very successful in explaining different features ofbehavior and used in problem solving although they little to dobehavior and used in problem solving although they little to do

with brainwith brain--like information processing.like information processing.


9/30

Motivation: practicalMotivation: practical

Large computer power allows for buildingLarge computer power allows for building

AI and CI has not been able to create decent humanAI and CI has not been able to create decent human--computer interfaces, solve problems in computer vision,computer interfaces, solve problems in computer vision,natural language understanding, cognitive search and datanatural language understanding, cognitive search and data

mining, or even reasoning in theorem proving.mining, or even reasoning in theorem proving.

Practical: humanized, cognitive computer applicationsPractical: humanized, cognitive computer applicationsrequire a brainrequire a brain--like architecture (either software orlike architecture (either software or

hardware) to deal with such problems efficiently; it is at thehardware) to deal with such problems efficiently; it is at thecenter of cognitive robotics.center of cognitive robotics.


10/30

Some failed attempts

Many have proposed the construction of brainMany have proposed the construction of brain--like computers,like computers,frequently using special hardware.frequently using special hardware.

Connection Machines from Thinking Machines, Inc. (D. Hills,Connection Machines from Thinking Machines, Inc. (D. Hills,1987) was commercially almost successful, but never become1987) was commercially almost successful, but never become

massively parallel and the company went bankrupt.massively parallel and the company went bankrupt.

CAM Brain (ATR Kyoto)CAM Brain (ATR Kyoto) failed attempt to evolve the largefailed attempt to evolve the large--scale cellular neural network; based on a bad idea that one canscale cellular neural network; based on a bad idea that one canevolve functions without knowing them. It is impossible toevolve functions without knowing them. It is impossible to

repeat evolutionary process (lack of data about initial organismsrepeat evolutionary process (lack of data about initial organismsand environment, almost infinite number of evolutionaryand environment, almost infinite number of evolutionarypathways). Evolutionary algorithms require supervision (fitnesspathways). Evolutionary algorithms require supervision (fitnessfunction) but it is not clear how to create fitness functions forfunction) but it is not clear how to create fitness functions forparticular brain structures without knowing their functions first;particular brain structures without knowing their functions first;but if we know the function we can program it without evolving.but if we know the function we can program it without evolving.


11/30

Special hardware?

Many have proposed the construction of brainMany have proposed the construction of brain--like computers,like computers,frequently using special hardware, but there are no largefrequently using special hardware, but there are no large--scalescaleconstructions so far.constructions so far.

Needed: elements based on spiking biological neurons and theNeeded: elements based on spiking biological neurons and the

layered 2layered 2--D anatomy of mammalian cerebral cortex.D anatomy of mammalian cerebral cortex.

ALAVLSI, AttendALAVLSI, Attend--toto--learn and learnlearn and learn--toto--attend with analog VLSI, EUattend with analog VLSI, EUIST Consortium 2002IST Consortium 2002--2005, Plymouth, ETH, Uni Berne, Siemens.2005, Plymouth, ETH, Uni Berne, Siemens.A general architecture for perceptual attention and learning based onA general architecture for perceptual attention and learning based on

neuromorphic VLSI technology.neuromorphic VLSI technology.

Coherent motion + speech categorization, project ends in 2005.Coherent motion + speech categorization, project ends in 2005.

PP--RAM neurons, KCL?RAM neurons, KCL?


12/30

Natural perception

Spectrogram of speech: hearing a sentence.Spectrogram of speech: hearing a sentence.


13/30

Spiking vs. mean field

Brain: 10Brain: 101111 NeuronsNeurons

Neuron PoolsNeuron Pools

0

( , )( ) lim

spikes

t

n t t t A t

M t( p

(!

(

Pool Activity:

Pool

M Neurons

neuron

12

3

M

spikest(

t

Linked PoolsLinked Pools

A

C

B

( ) ...B B ex

I I aF A I Xx

!

x

Mean-Field

Model:

F

neuron 1neuron 1

neuron 2neuron 2

Networks of Spiking NeuronsNetworks of Spiking Neurons

)())(()( tIVtVgtVdt

d

synLimim!X

2 ( )V t

t

Integrateand

FireModel:


14/30

,

, ,

,, ,

, ,

, 2

, ,

( ) ( ( ) ) ( )

( ) ( ( ) ) ( )

( ( ) )( ) ( )

(1 [ ]exp( 0.062 ( ) /3 .57))

( )

AMPA ext

AMPA ext AMPA ext i E ij j

j

AMPA recAMPA rec AMPA rec i E ij j

j

NMDA rec i E NMDA rec

NMDA rec ij j

ji

GABA rec GABA

I t g V t V w s t

I t g V t V w s t

g V t V I t w s t

Mg V t

I t g

!

!

!

!

,( ( ) ) ( )GABA recrec i E ij jj

V t V w s t

,

,

( )( ) ( )

( )( ) ( )(1 ( ))

( )( ) ( )

( )( ) ( )

AMPA

jAMPA k

j j

kAMPA

NMDA

jNMDA NMDA

j j j

NMDA decay

NMDA

jNMDA k

j j

kNMDA rise

GABA

jGABA k

j j

kGABA

s tds t t t

dt

s tds t x t s t

dt

x tdx t t t

dt

s tds t t t

dt

HX

EX

HX

HX

!

!

!

!

Synaptic Dynamics

EPSP, IPSP)(tIsyn

Spike

Spike

SomaSynapses

synCmC mR

synR


15/30

Darwin/Nomad robots

G. EdelmanG. Edelman (Neurosciences Institute) & collaborators, created a series(Neurosciences Institute) & collaborators, created a seriesof Darwin automata, brainof Darwin automata, brain--based devices, physical devices whosebased devices, physical devices whosebehavior is controlled by a simulated nervous system.behavior is controlled by a simulated nervous system.

(i)(i) The device must engage in a behavioral task.The device must engage in a behavioral task.

(ii)(ii) The devices behavior must be controlled by a simulatedThe devices behavior must be controlled by a simulatednervous system having a design that reflects the brainsnervous system having a design that reflects the brainsarchitecture and dynamics.architecture and dynamics.

(iii)(iii) The devices behavior is modified by a reward or value system thatThe devices behavior is modified by a reward or value system thatsignals the salience of environmental cues to its nervous system.signals the salience of environmental cues to its nervous system.

(iv)(iv) The device must be situated in the real world.The device must be situated in the real world.

Darwin VII consists of: a mobile base equipped with a CCD camera andDarwin VII consists of: a mobile base equipped with a CCD camera andIR sensor for vision, microphones for hearing, conductivity sensors forIR sensor for vision, microphones for hearing, conductivity sensors fortaste, and effectors for movement of its base, of its head, and of ataste, and effectors for movement of its base, of its head, and of agripping manipulator having one degreegripping manipulator having one degree--ofof--freedom; 53K mean firingfreedom; 53K mean firing

+phase neurons, 1.7 M synapses, 28 brain areas.+phase neurons, 1.7 M synapses, 28 brain areas.


16/30

Blue Brain

The Blue Brain Project was launched by the Brain Mind Institute, EPFL,The Blue Brain Project was launched by the Brain Mind Institute, EPFL,Switzerland and IBM, USA in MaySwitzerland and IBM, USA in May0505, now, now over 120'000over 120'000 WWW pagesWWW pages..

The EPFL Blue Gene is the 8th fastest supercomputer in the worldThe EPFL Blue Gene is the 8th fastest supercomputer in the world..

CCan simulate about 100an simulate about 100MM minimal compartment neurons or 10minimal compartment neurons or 10--50'00050'000multimulti--compartmental neuronscompartmental neurons,, with 10with 1033--101044 xx more synapses.more synapses. NNextext

generationgeneration BG willBG will simulatesimulate >10>1099 neurons with significant complexity.neurons with significant complexity.

FFirst objective is to create a cellular level, software replica of theirst objective is to create a cellular level, software replica of theNeocortical Column for realNeocortical Column for real--time simulations.time simulations.

The Blue Brain Project will soon invite researchers to build their ownThe Blue Brain Project will soon invite researchers to build their own

models of different brain regions in different species and at differentmodels of different brain regions in different species and at differentlevels of detail using Blue Brain Software for simulation on Blue Gene.levels of detail using Blue Brain Software for simulation on Blue Gene.These models will be deposited in an Internet Database from whichThese models will be deposited in an Internet Database from whichBlue Brain software can extract and connect models together to buildBlue Brain software can extract and connect models together to buildbrain regions and begin the first whole brain simulations.brain regions and begin the first whole brain simulations.


17/30

Blue Brain 2

Models at different level of complexity:Models at different level of complexity:http://bluebrainproject.epfl.ch/http://bluebrainproject.epfl.ch/

1. The Blue Synapse: A molecular level model of a single synapse1. The Blue Synapse: A molecular level model of a single synapse..

2. The Blue Neuron: A molecular level model of a single neuron2. The Blue Neuron: A molecular level model of a single neuron..

3. The Blue Column: A cellular level model of the Neocortical column3. The Blue Column: A cellular level model of the Neocortical columnwith 10with 10KK neuronsneurons,, laterlater 50K, 100M connections.50K, 100M connections.

4. The Blue Neocortex: A simplified Blue Column will be duplicated to4. The Blue Neocortex: A simplified Blue Column will be duplicated toproduce Neocortical regions and eventually and entire Neocortex.produce Neocortical regions and eventually and entire Neocortex.

5. The Blue Brain Project will also build models of other Cortical and5. The Blue Brain Project will also build models of other Cortical andSubcortical models of the brainSubcortical models of the brain,, and sensory + motor organsand sensory + motor organs..


18/30

Blue Column

A detailed and faithful computer reproduction of the Neocortical Column.A detailed and faithful computer reproduction of the Neocortical Column.

It will first be based on the data obtained from rat somatosensory cortexIt will first be based on the data obtained from rat somatosensory cortexat 2 weeks of age. Once built and calibrated with iterative simulationsat 2 weeks of age. Once built and calibrated with iterative simulationsand experiments, comparative data will be used to build columns inand experiments, comparative data will be used to build columns in

different brain regions, ages and species, including humans.different brain regions, ages and species, including humans.

BC will be composed of 10BC will be composed of 1044 morphologically complex neurons withmorphologically complex neurons withactive ionic channels, interconnected in a 3active ionic channels, interconnected in a 3--dimensional (3D) space withdimensional (3D) space with101077 --101088 dynamic synapses, receiving 10dynamic synapses, receiving 1033 --101044 external input synapses,external input synapses,

generating 10generating 1033

--101044

external output synapses.external output synapses.

Neurons use dynamic and stochastic synaptic transmission rules forNeurons use dynamic and stochastic synaptic transmission rules forlearning, with metalearning, with meta--plasticity, supervised & reward learning algorithmsplasticity, supervised & reward learning algorithmsfor all synapses.for all synapses.


19/30

Blue Column 3

Project will include creation of:Project will include creation of:

Databases: NOBASE holds 3D reconstructed model neurons,Databases: NOBASE holds 3D reconstructed model neurons,synapses, synaptic pathways, microcircuit statistics, computer modelsynapses, synaptic pathways, microcircuit statistics, computer modelneurons, virtual neurons.neurons, virtual neurons.

Visualization: BlueBuilder, BlueVision and BlueAnalsysis. 2D, 3DVisualization: BlueBuilder, BlueVision and BlueAnalsysis. 2D, 3Dand immersive visualization systems are being developed.and immersive visualization systems are being developed.

Simulation: a simulation environment for large scale simulations ofSimulation: a simulation environment for large scale simulations ofmorphologically complex neurons on 8000 processors of IBM's Bluemorphologically complex neurons on 8000 processors of IBM's BlueGene supercomputer.Gene supercomputer.

Simulations & experiments: iterations between large scaleSimulations & experiments: iterations between large scalesimulations of neocortical microcircuits and experiments in order tosimulations of neocortical microcircuits and experiments in order toverify the computational model and explore predictions.verify the computational model and explore predictions.

Verification: in vivo = in silico?Verification: in vivo = in silico?


20/30

CCortex

Artificial Development (www.ad.com) is buildingArtificial Development (www.ad.com) is building CCortexCCortex,,a complete 20G neuron 20T connection simulation of thea complete 20G neuron 20T connection simulation of theHuman Cortex and peripheral systems, on a cluster of 500Human Cortex and peripheral systems, on a cluster of 500computerscomputers -- the largest neural network crated to date.the largest neural network crated to date.

Artificial Development plans to deliver a wide range of commercialArtificial Development plans to deliver a wide range of commercialproducts based on artificial versions of the human brain that will enhanceproducts based on artificial versions of the human brain that will enhancebusiness relationships globally.business relationships globally.

Rather unlikely? Simulation of PentiumRather unlikely? Simulation of Pentium

Not much has changed in the last year on their web pageNot much has changed in the last year on their web page,, except thatexcept thatAD opened a lab in Kochi, Kerala, India, to uncover relevant informationAD opened a lab in Kochi, Kerala, India, to uncover relevant informationon the functioning on the human brain, and help model and interpret theon the functioning on the human brain, and help model and interpret thedata. The company is run by Marcosdata. The company is run by Marcos GuillenGuillen, who made money as ISP, who made money as ISPin Spain but has no experience in neuroscience or simulations.in Spain but has no experience in neuroscience or simulations.


21/30

The Ersatz Brain ProjectThe Ersatz Brain Project

Vision: in 2050 the personal computer you buy in WalVision: in 2050 the personal computer you buy in Wal--Mart will have two CPUsMart will have two CPUswith very different architecture:with very different architecture:

First, a traditional von Neumann machine that runs spreadsheets, doesFirst, a traditional von Neumann machine that runs spreadsheets, doesword processing, keeps your calendar straight, etc. etc.word processing, keeps your calendar straight, etc. etc.

Second, a brainSecond, a brain--like chiplike chip To handle the interface with the von Neumann machine, To handle the interface with the von Neumann machine,

Give you the data that you need from the Web or your files.Give you the data that you need from the Web or your files.

Be your silicon friend, guide, and confidant.Be your silicon friend, guide, and confidant.

Project based on modeling of cortical columns of various sizesProject based on modeling of cortical columns of various sizes(minicolumns ~10(minicolumns ~1022, plain ~10, plain ~1044, and hypercolumns ~10, and hypercolumns ~1055), sparsely), sparselyconnected (0.001% in the brain).connected (0.001% in the brain).

NofN, Network of Networks approximation using 2D BSB (Brain in a Box)NofN, Network of Networks approximation using 2D BSB (Brain in a Box)network, similar in design to Connection Machines, but more processors.network, similar in design to Connection Machines, but more processors.


22/30

Conscious machines: HaikonenConscious machines: HaikonenConscious machines: HaikonenConscious machines: Haikonen

Haikonen has done some simulations based on a rather straightforwardHaikonen has done some simulations based on a rather straightforwarddesign, with neural models feeding the sensory information (with WTAdesign, with neural models feeding the sensory information (with WTAassociative memory) into the associative working memory circuits.associative memory) into the associative working memory circuits.


23/30

by Tetsuya HoyaBSI-RIKEN, Japan

Lab. Advanced Brain Signal Processing

Artificial Mind System (AMS)Artificial Mind System (AMS)Kernel Memory ApproachKernel Memory Approach

Series: Studies in ComputationalIntelligence (SCI), Vol. 1 (270p)

Springer-Verlag: HeidelbergAug. 2005

available from:http://www.springeronline.com/


24/30

To provide an engineering account to modelvarious functionalities related to mind, motivatedfrom the modularity principle of mind (Fodor,1983; Hobson, 1999).

To embody each module and their mutual dataprocessing within the AMS, by means of a newconnectionist model, kernel memory.

Thereby, to develop a new form ofartificial

intelligent system with ideas from a broaderspectrum of brain scientific studies artificialintelligence, cognitive science/psychology,connectionism, consciousness studies, generalneuroscience, linguistics, patternrecognition/data clustering, robotics, and signal

processing.

Objectives:

Artificial Mind System (AMS)Artificial Mind System (AMS)Kernel Memory ApproachKernel Memory Approach


25/30

Machine consciousness: OwenMachine consciousness: Owen

Holland Owen, ExeterHolland Owen, Exeterhttp://www.machineconsciousness.org/http://www.machineconsciousness.org/

Owen Holland at the University of Essex and Tom Troscianko and IanOwen Holland at the University of Essex and Tom Troscianko and IanGilchrist at the University of Bristol, have received 493,000 (714,000Gilchrist at the University of Bristol, have received 493,000 (714,000Euros, or $833,000) from the Eng. & Phys. Sci. Res. Council for aEuros, or $833,000) from the Eng. & Phys. Sci. Res. Council for a

project 'Machine consciousness through internal modeling, 2004project 'Machine consciousness through internal modeling, 2004--2007.2007.

To survive robots will plan actions, build a model of the world and aTo survive robots will plan actions, build a model of the world and amodel of itselfmodel of itself -- its body, sensors, manipulators, preferences, history its body, sensors, manipulators, preferences, history Biological vision systems is the basis for internal processes and modelsBiological vision systems is the basis for internal processes and modelsand will be accessible to the investigating team as visual displays. Theand will be accessible to the investigating team as visual displays. Themain focus of interest will be the selfmain focus of interest will be the self--model; its characteristics andmodel; its characteristics andinternal changes are expected to resemble those of the conscious selfinternal changes are expected to resemble those of the conscious selfin humans, perhaps closely enough to enable some of the robots to bein humans, perhaps closely enough to enable some of the robots to beregarded as possessing a form of machine consciousness.regarded as possessing a form of machine consciousness.Increasingly complex biologically inspired autonomous mobile robotsIncreasingly complex biologically inspired autonomous mobile robotsforced to survive in a series of progressively more difficult environments,forced to survive in a series of progressively more difficult environments,and will then study the external and internal behavior of the robots,and will then study the external and internal behavior of the robots,

looking for signs and characteristics of consciousness.looking for signs and characteristics of consciousness.


26/30

Bayesian Confidence Propagating NN.Bayesian Confidence Propagating NN.

Johansson/Lansner ideas:Johansson/Lansner ideas:

oo Assumption: functional principles of cortex reside on a much higherAssumption: functional principles of cortex reside on a much higherlevel of abstraction than that of the single neuron i.e. closer tolevel of abstraction than that of the single neuron i.e. closer toabstractions like ANN and connectionist models.abstractions like ANN and connectionist models.

oo Target: artificial brain, compact, lowTarget: artificial brain, compact, low--power, multipower, multi--network NN.network NN.

oo Mapping of cortical structure onto the BCPNN, an attractor network.Mapping of cortical structure onto the BCPNN, an attractor network.oo Implementation of BCPNN based on hyper columnar modules.Implementation of BCPNN based on hyper columnar modules.

oo Hypercolumn needs 5Hypercolumn needs 5..101099 ops, with about 2ops, with about 2..101066 hypercolumns inhypercolumns inhuman cortex, giving about 10human cortex, giving about 101616 ops.ops.

oo No detailed structure proposed.No detailed structure proposed.


27/30

Intelligent Distributed Agents.Intelligent Distributed Agents.

Stan Franklin (Memphis): IDA is an intelligent, autonomous softwareStan Franklin (Memphis): IDA is an intelligent, autonomous softwareagent that does personnel work for the US Navy.agent that does personnel work for the US Navy.


28/30

IDA insideIDA inside

Based on Baars Global Workspace theory.Based on Baars Global Workspace theory.


29/30

IDA in actionIDA in action


30/30

Hal Baby BrainHal Baby Brain..

Evolve language:Evolve language: www.awww.a--i.comi.comSo far: simple 2So far: simple 2--3 words but meaningful.3 words but meaningful.

Will it ever make it to higher level? Doubtful.Will it ever make it to higher level? Doubtful.

Date post:	08-Apr-2018
Category:	Documents
Upload:	sikandar-khan
View:	213 times
Download:	0 times

05-09-BrainBuilding-review

Documents