SlideShare a Scribd company logo
ď‚ž Unsupervised ANN
ď‚ž Stability-Plasticity Dilemma
ď‚ž Adaptive Resonance Theory basics
ď‚ž ART Architecture
ď‚ž Algorithm
ď‚ž Types of ART NN
ď‚ž Applications
ď‚ž References
ď‚ž Usually 2-layer ANN
ď‚ž Only input data are given
ď‚ž ANN must self-organise
output
 Two main models: Kohonen’s
SOM and Grossberg’s ART
ď‚ž Clustering applications
Output layer
Feature layer
Every learning system faces the plasticity-stability dilemma.
ď‚ž The plasticity-stability dilemma poses few questions :
ART stands for "Adaptive Resonance Theory", invented by Stephen
Grossberg in 1976.
ď‚žART represents a family of neural networks.
ď‚žThe basic ART System is an unsupervised learning model.
ď‚žThe term "resonance" refers to resonant state of a neural network in
which a category prototype vector matches close enough to the current
input vector. ART matching leads to this resonant state, which permits
learning. The network learns only in its resonant state.
The key innovation of ART is the use of
“expectations.”
› As each input is presented to the network, it is compared
with the prototype vector that is most closely matches
(the expectation).
› If the match between the prototype and the input vector is
NOT adequate, a new prototype is selected. In this way,
previous learned memories (prototypes) are not eroded by
new learning.
Input
Layer 1
(Retina)
Layer 2
(Visual Cortex)
LTM
(Adaptive
Weights)
STM
Normalization Constrast
Enhancement
Basic ART architecture
Grossberg competitive network
ď‚ž The L1-L2 connections are instars, which performs a clustering
(or categorization) operation. When an input pattern is presented,
it is multiplied (after normalization) by the L1-L2 weight matrix.
ď‚ž A competition is performed at Layer 2 to determine which row of
the weight matrix is closest to the input vector. That row is then
moved toward the input vector.
ď‚ž After learning is complete, each row of the L1-L2 weight matrix is
a prototype pattern, which represents a cluster (or a category) of
input vectors.
ď‚ž Learning of ART networks also occurs in a set of feedback
connections from Layer 2 to Layer 1. These connections are
outstars which perform pattern recall.
ď‚ž When a node in Layer 2 is activated, this reproduces a
prototype pattern (the expectation) at layer 1.
ď‚ž Layer 1 then performs a comparison between the
expectation and the input pattern.
ď‚ž When the expectation and the input pattern are NOT closely
matched, the orienting subsystem causes a reset in Layer 2.
ď‚ž The reset disables the current winning neuron, and the
current expectation is removed.
ď‚ž A new competition is then performed in Layer 2, while the
previous winning neuron is disable.
ď‚ž The new winning neuron in Layer 2 projects a new
expectation to Layer 1, through the L2-L1 connections.
ď‚ž This process continues until the L2-L1 expectation provides a
close enough match to the input pattern.
ď‚ž Bottom-up weights bij
ď‚ž Top-down weights tij
› Store class template
ď‚ž Input nodes
› Vigilance test
› Input normalisation
ď‚ž Output nodes
› Forward matching
ď‚ž Long-term memory
› ANN weights
ď‚ž Short-term memory
› ANN activation pattern top down
bottom up (normalised)
ď‚ž The basic ART system is unsupervised learning
model. It typically consists of
› a comparison field and a recognition field composed of
neurons,
› a vigilance parameter, and
› a reset module
ď‚ž Comparison field
› The comparison field takes an input vector (a one-dimensional array of
values) and transfers it to its best match in the recognition field. Its best match
is the single neuron whose set of weights (weight vector) most closely
matches the input vector.
ď‚ž Recognition field
› Each recognition field neuron, outputs a negative signal proportional to that
neuron's quality of match to the input vector to each of the other recognition
field neurons and inhibits their output accordingly. In this way the recognition
field exhibits lateral inhibition, allowing each neuron in it to represent a
category to which input vectors are classified.
ď‚ž Vigilance parameter
› After the input vector is classified, a reset module compares the
strength of the recognition match to a vigilance parameter. The
vigilance parameter has considerable influence on the system.
ď‚ž Reset Module
› The reset module compares the strength of the recognition match to
the vigilance parameter.
› If the vigilance threshold is met, then training commences.
Adapt winner
node
Initialise uncommitted
node
new pattern
categorisation
known unknown
recognition
comparison
ď‚ž Incoming pattern matched with
stored cluster templates
ď‚ž If close enough to stored template
joins best matching cluster,
weights adapted
ď‚ž If not, a new cluster is initialised
with pattern as template
ď‚ž ART-1
› Binary input vectors
› Unsupervised NN that can be complemented with external
changes to the vigilance parameter
ď‚ž ART-2
› Real-valued input vectors
ď‚ž ART-3
› Parallel search of compressed or distributed pattern
recognition codes in a NN hierarchy.
› Search process leads to the discovery of appropriate
representations of a non stationary input environment.
› Chemical properties of the synapse emulated in the search
process
1 2 3
1 2 3 4Input
layer
Output layer
with inhibitory
connections
),( 3,44,3 tb
The ART-1 Network
• Mobile robot control
• Facial recognition
• Land cover classification
• Target recognition
• Medical diagnosis
• Signature verification
 S. Rajasekaran, G.A.V. Pai, “Neural Networks, Fuzzy Logic and
Genetic Algorithms”, Prentice Hall of India, Adaptive Resonance
Theory, Chapter 5.
 Jacek M. Zurada, “Introduction to Artificial Neural Systems”, West
Publishing Company, Matching & Self organizing maps, Chapter 7.
ď‚ž Adaptive Resonance Theory, Soft computing lecture notes,
“http://www.myreaders.info/html/soft_computing.html”
@ chd.naveen@gmail.com
/chd.naveen
@saini_naveen87
/NaveenKumar11
www.elixir-india.com
Adaptive Resonance Theory

More Related Content

What's hot

Back propagation
Back propagationBack propagation
Back propagation
Nagarajan
 
Perceptron (neural network)
Perceptron (neural network)Perceptron (neural network)
Perceptron (neural network)
EdutechLearners
 
Feed forward ,back propagation,gradient descent
Feed forward ,back propagation,gradient descentFeed forward ,back propagation,gradient descent
Feed forward ,back propagation,gradient descent
Muhammad Rasel
 
Kohonen self organizing maps
Kohonen self organizing mapsKohonen self organizing maps
Kohonen self organizing maps
raphaelkiminya
 
Max net
Max netMax net
Perceptron & Neural Networks
Perceptron & Neural NetworksPerceptron & Neural Networks
Perceptron & Neural Networks
NAGUR SHAREEF SHAIK
 
Problem reduction AND OR GRAPH & AO* algorithm.ppt
Problem reduction AND OR GRAPH & AO* algorithm.pptProblem reduction AND OR GRAPH & AO* algorithm.ppt
Problem reduction AND OR GRAPH & AO* algorithm.ppt
arunsingh660
 
backpropagation in neural networks
backpropagation in neural networksbackpropagation in neural networks
backpropagation in neural networks
Akash Goel
 
Counterpropagation NETWORK
Counterpropagation NETWORKCounterpropagation NETWORK
Counterpropagation NETWORKESCOM
 
Mc culloch pitts neuron
Mc culloch pitts neuronMc culloch pitts neuron
Associative memory network
Associative memory networkAssociative memory network
Associative memory network
Dr. C.V. Suresh Babu
 
Learning Methods in a Neural Network
Learning Methods in a Neural NetworkLearning Methods in a Neural Network
Learning Methods in a Neural Network
Saransh Choudhary
 
Artificial Neural Networks Lect3: Neural Network Learning rules
Artificial Neural Networks Lect3: Neural Network Learning rulesArtificial Neural Networks Lect3: Neural Network Learning rules
Artificial Neural Networks Lect3: Neural Network Learning rules
Mohammed Bennamoun
 
Feedforward neural network
Feedforward neural networkFeedforward neural network
Feedforward neural network
Sopheaktra YONG
 
Multilayer & Back propagation algorithm
Multilayer & Back propagation algorithmMultilayer & Back propagation algorithm
Multilayer & Back propagation algorithm
swapnac12
 
Self-organizing map
Self-organizing mapSelf-organizing map
Self-organizing map
Tarat Diloksawatdikul
 
distributed shared memory
 distributed shared memory distributed shared memory
distributed shared memoryAshish Kumar
 
Hebb network
Hebb networkHebb network
I. AO* SEARCH ALGORITHM
I. AO* SEARCH ALGORITHMI. AO* SEARCH ALGORITHM
I. AO* SEARCH ALGORITHM
vikas dhakane
 
Activation function
Activation functionActivation function
Activation function
Astha Jain
 

What's hot (20)

Back propagation
Back propagationBack propagation
Back propagation
 
Perceptron (neural network)
Perceptron (neural network)Perceptron (neural network)
Perceptron (neural network)
 
Feed forward ,back propagation,gradient descent
Feed forward ,back propagation,gradient descentFeed forward ,back propagation,gradient descent
Feed forward ,back propagation,gradient descent
 
Kohonen self organizing maps
Kohonen self organizing mapsKohonen self organizing maps
Kohonen self organizing maps
 
Max net
Max netMax net
Max net
 
Perceptron & Neural Networks
Perceptron & Neural NetworksPerceptron & Neural Networks
Perceptron & Neural Networks
 
Problem reduction AND OR GRAPH & AO* algorithm.ppt
Problem reduction AND OR GRAPH & AO* algorithm.pptProblem reduction AND OR GRAPH & AO* algorithm.ppt
Problem reduction AND OR GRAPH & AO* algorithm.ppt
 
backpropagation in neural networks
backpropagation in neural networksbackpropagation in neural networks
backpropagation in neural networks
 
Counterpropagation NETWORK
Counterpropagation NETWORKCounterpropagation NETWORK
Counterpropagation NETWORK
 
Mc culloch pitts neuron
Mc culloch pitts neuronMc culloch pitts neuron
Mc culloch pitts neuron
 
Associative memory network
Associative memory networkAssociative memory network
Associative memory network
 
Learning Methods in a Neural Network
Learning Methods in a Neural NetworkLearning Methods in a Neural Network
Learning Methods in a Neural Network
 
Artificial Neural Networks Lect3: Neural Network Learning rules
Artificial Neural Networks Lect3: Neural Network Learning rulesArtificial Neural Networks Lect3: Neural Network Learning rules
Artificial Neural Networks Lect3: Neural Network Learning rules
 
Feedforward neural network
Feedforward neural networkFeedforward neural network
Feedforward neural network
 
Multilayer & Back propagation algorithm
Multilayer & Back propagation algorithmMultilayer & Back propagation algorithm
Multilayer & Back propagation algorithm
 
Self-organizing map
Self-organizing mapSelf-organizing map
Self-organizing map
 
distributed shared memory
 distributed shared memory distributed shared memory
distributed shared memory
 
Hebb network
Hebb networkHebb network
Hebb network
 
I. AO* SEARCH ALGORITHM
I. AO* SEARCH ALGORITHMI. AO* SEARCH ALGORITHM
I. AO* SEARCH ALGORITHM
 
Activation function
Activation functionActivation function
Activation function
 

Similar to Adaptive Resonance Theory

Neural network final NWU 4.3 Graphics Course
Neural network final NWU 4.3 Graphics CourseNeural network final NWU 4.3 Graphics Course
Neural network final NWU 4.3 Graphics Course
Mohaiminur Rahman
 
ai7.ppt
ai7.pptai7.ppt
ai7.ppt
MrHacker61
 
ANNs have been widely used in various domains for: Pattern recognition Funct...
ANNs have been widely used in various domains for: Pattern recognition  Funct...ANNs have been widely used in various domains for: Pattern recognition  Funct...
ANNs have been widely used in various domains for: Pattern recognition Funct...
vijaym148
 
Artificial neural network
Artificial neural networkArtificial neural network
Artificial neural network
sweetysweety8
 
Backpropagation.pptx
Backpropagation.pptxBackpropagation.pptx
Backpropagation.pptx
VandanaVipparthi
 
Introduction Of Artificial neural network
Introduction Of Artificial neural networkIntroduction Of Artificial neural network
Introduction Of Artificial neural network
Nagarajan
 
ai7.ppt
ai7.pptai7.ppt
ai7.ppt
qwerty432737
 
Neuralnetwork 101222074552-phpapp02
Neuralnetwork 101222074552-phpapp02Neuralnetwork 101222074552-phpapp02
Neuralnetwork 101222074552-phpapp02Deepu Gupta
 
Classification by back propagation, multi layered feed forward neural network...
Classification by back propagation, multi layered feed forward neural network...Classification by back propagation, multi layered feed forward neural network...
Classification by back propagation, multi layered feed forward neural network...
bihira aggrey
 
Adaptive equalization
Adaptive equalizationAdaptive equalization
Adaptive equalization
Kamal Bhatt
 
neural-networks (1)
neural-networks (1)neural-networks (1)
neural-networks (1)rockeysuseelan
 
Neural Networks
Neural NetworksNeural Networks
Neural Networks
NikitaRuhela
 
SoftComputing6
SoftComputing6SoftComputing6
SoftComputing6
DrPrafullNarooka
 
Terminology Machine Learning
Terminology Machine LearningTerminology Machine Learning
Terminology Machine Learning
DataminingTools Inc
 
Echo state networks and locomotion patterns
Echo state networks and locomotion patternsEcho state networks and locomotion patterns
Echo state networks and locomotion patterns
Vito Strano
 
Adapt-Resona-Theory, adaptive resonnance theorey
Adapt-Resona-Theory, adaptive resonnance theoreyAdapt-Resona-Theory, adaptive resonnance theorey
Adapt-Resona-Theory, adaptive resonnance theorey
NaveenKumar5162
 
Artifical Neural Network
Artifical Neural NetworkArtifical Neural Network
Artifical Neural Network
mahalakshmimalini
 
Artificial Neural Networks ppt.pptx for final sem cse
Artificial Neural Networks  ppt.pptx for final sem cseArtificial Neural Networks  ppt.pptx for final sem cse
Artificial Neural Networks ppt.pptx for final sem cse
NaveenBhajantri1
 
Hardware Implementation of Spiking Neural Network (SNN)
Hardware Implementation of Spiking Neural Network (SNN)Hardware Implementation of Spiking Neural Network (SNN)
Hardware Implementation of Spiking Neural Network (SNN)
supratikmondal6
 

Similar to Adaptive Resonance Theory (20)

Neural network final NWU 4.3 Graphics Course
Neural network final NWU 4.3 Graphics CourseNeural network final NWU 4.3 Graphics Course
Neural network final NWU 4.3 Graphics Course
 
ai7.ppt
ai7.pptai7.ppt
ai7.ppt
 
ANNs have been widely used in various domains for: Pattern recognition Funct...
ANNs have been widely used in various domains for: Pattern recognition  Funct...ANNs have been widely used in various domains for: Pattern recognition  Funct...
ANNs have been widely used in various domains for: Pattern recognition Funct...
 
Artificial neural network
Artificial neural networkArtificial neural network
Artificial neural network
 
Backpropagation.pptx
Backpropagation.pptxBackpropagation.pptx
Backpropagation.pptx
 
Introduction Of Artificial neural network
Introduction Of Artificial neural networkIntroduction Of Artificial neural network
Introduction Of Artificial neural network
 
ai7.ppt
ai7.pptai7.ppt
ai7.ppt
 
Neuralnetwork 101222074552-phpapp02
Neuralnetwork 101222074552-phpapp02Neuralnetwork 101222074552-phpapp02
Neuralnetwork 101222074552-phpapp02
 
Classification by back propagation, multi layered feed forward neural network...
Classification by back propagation, multi layered feed forward neural network...Classification by back propagation, multi layered feed forward neural network...
Classification by back propagation, multi layered feed forward neural network...
 
Adaptive equalization
Adaptive equalizationAdaptive equalization
Adaptive equalization
 
neural-networks (1)
neural-networks (1)neural-networks (1)
neural-networks (1)
 
Neural Networks
Neural NetworksNeural Networks
Neural Networks
 
SoftComputing6
SoftComputing6SoftComputing6
SoftComputing6
 
Terminology Machine Learning
Terminology Machine LearningTerminology Machine Learning
Terminology Machine Learning
 
Echo state networks and locomotion patterns
Echo state networks and locomotion patternsEcho state networks and locomotion patterns
Echo state networks and locomotion patterns
 
Adapt-Resona-Theory, adaptive resonnance theorey
Adapt-Resona-Theory, adaptive resonnance theoreyAdapt-Resona-Theory, adaptive resonnance theorey
Adapt-Resona-Theory, adaptive resonnance theorey
 
Artifical Neural Network
Artifical Neural NetworkArtifical Neural Network
Artifical Neural Network
 
20120140503023
2012014050302320120140503023
20120140503023
 
Artificial Neural Networks ppt.pptx for final sem cse
Artificial Neural Networks  ppt.pptx for final sem cseArtificial Neural Networks  ppt.pptx for final sem cse
Artificial Neural Networks ppt.pptx for final sem cse
 
Hardware Implementation of Spiking Neural Network (SNN)
Hardware Implementation of Spiking Neural Network (SNN)Hardware Implementation of Spiking Neural Network (SNN)
Hardware Implementation of Spiking Neural Network (SNN)
 

More from Naveen Kumar

Security in GSM(2G) and UMTS(3G) Networks
Security in GSM(2G) and UMTS(3G) NetworksSecurity in GSM(2G) and UMTS(3G) Networks
Security in GSM(2G) and UMTS(3G) Networks
Naveen Kumar
 
Mobile tower radiation
Mobile tower radiationMobile tower radiation
Mobile tower radiation
Naveen Kumar
 
Mobile security
Mobile securityMobile security
Mobile security
Naveen Kumar
 
Ph.D Research proposal
Ph.D Research proposalPh.D Research proposal
Ph.D Research proposal
Naveen Kumar
 
Wi-Fi Technology
Wi-Fi TechnologyWi-Fi Technology
Wi-Fi Technology
Naveen Kumar
 
Cell Phone Antennas
Cell Phone AntennasCell Phone Antennas
Cell Phone Antennas
Naveen Kumar
 
Thesis on PIFA
Thesis on PIFAThesis on PIFA
Thesis on PIFA
Naveen Kumar
 
Electronics Quiz
Electronics QuizElectronics Quiz
Electronics Quiz
Naveen Kumar
 
VHDL coding in Xilinx
VHDL coding in XilinxVHDL coding in Xilinx
VHDL coding in Xilinx
Naveen Kumar
 
Optimization in HFSS
Optimization in HFSSOptimization in HFSS
Optimization in HFSS
Naveen Kumar
 
Free space optical communication
Free space optical communicationFree space optical communication
Free space optical communication
Naveen Kumar
 
A Multi-Band PIFA with Slotted Ground Plane
A Multi-Band PIFA with Slotted Ground Plane A Multi-Band PIFA with Slotted Ground Plane
A Multi-Band PIFA with Slotted Ground Plane
Naveen Kumar
 
Study of Planar Inverted - F Antenna (PIFA) for mobile devices
Study of Planar Inverted - F Antenna (PIFA) for mobile devices Study of Planar Inverted - F Antenna (PIFA) for mobile devices
Study of Planar Inverted - F Antenna (PIFA) for mobile devices
Naveen Kumar
 
A novel low profile planar inverted f antenna (pifa) for mobile handsets
A novel low profile planar inverted f antenna (pifa) for mobile handsetsA novel low profile planar inverted f antenna (pifa) for mobile handsets
A novel low profile planar inverted f antenna (pifa) for mobile handsets
Naveen Kumar
 
A compact planar inverted-F antenna with slotted ground plane
A compact planar inverted-F antenna with slotted ground planeA compact planar inverted-F antenna with slotted ground plane
A compact planar inverted-F antenna with slotted ground plane
Naveen Kumar
 
Secure Socket Layer
Secure Socket LayerSecure Socket Layer
Secure Socket Layer
Naveen Kumar
 
UART
UARTUART
UART
Naveen Kumar
 
HDLC, PPP and SLIP
HDLC, PPP and SLIPHDLC, PPP and SLIP
HDLC, PPP and SLIP
Naveen Kumar
 
VLSI routing
VLSI routingVLSI routing
VLSI routing
Naveen Kumar
 

More from Naveen Kumar (20)

Security in GSM(2G) and UMTS(3G) Networks
Security in GSM(2G) and UMTS(3G) NetworksSecurity in GSM(2G) and UMTS(3G) Networks
Security in GSM(2G) and UMTS(3G) Networks
 
Mobile tower radiation
Mobile tower radiationMobile tower radiation
Mobile tower radiation
 
Mobile security
Mobile securityMobile security
Mobile security
 
Ph.D Research proposal
Ph.D Research proposalPh.D Research proposal
Ph.D Research proposal
 
Wi-Fi Technology
Wi-Fi TechnologyWi-Fi Technology
Wi-Fi Technology
 
Cell Phone Antennas
Cell Phone AntennasCell Phone Antennas
Cell Phone Antennas
 
Thesis on PIFA
Thesis on PIFAThesis on PIFA
Thesis on PIFA
 
Electronics Quiz
Electronics QuizElectronics Quiz
Electronics Quiz
 
VHDL coding in Xilinx
VHDL coding in XilinxVHDL coding in Xilinx
VHDL coding in Xilinx
 
Optimization in HFSS
Optimization in HFSSOptimization in HFSS
Optimization in HFSS
 
Free space optical communication
Free space optical communicationFree space optical communication
Free space optical communication
 
A Multi-Band PIFA with Slotted Ground Plane
A Multi-Band PIFA with Slotted Ground Plane A Multi-Band PIFA with Slotted Ground Plane
A Multi-Band PIFA with Slotted Ground Plane
 
Study of Planar Inverted - F Antenna (PIFA) for mobile devices
Study of Planar Inverted - F Antenna (PIFA) for mobile devices Study of Planar Inverted - F Antenna (PIFA) for mobile devices
Study of Planar Inverted - F Antenna (PIFA) for mobile devices
 
A novel low profile planar inverted f antenna (pifa) for mobile handsets
A novel low profile planar inverted f antenna (pifa) for mobile handsetsA novel low profile planar inverted f antenna (pifa) for mobile handsets
A novel low profile planar inverted f antenna (pifa) for mobile handsets
 
A compact planar inverted-F antenna with slotted ground plane
A compact planar inverted-F antenna with slotted ground planeA compact planar inverted-F antenna with slotted ground plane
A compact planar inverted-F antenna with slotted ground plane
 
Secure Socket Layer
Secure Socket LayerSecure Socket Layer
Secure Socket Layer
 
UART
UARTUART
UART
 
HDLC, PPP and SLIP
HDLC, PPP and SLIPHDLC, PPP and SLIP
HDLC, PPP and SLIP
 
AR model
AR modelAR model
AR model
 
VLSI routing
VLSI routingVLSI routing
VLSI routing
 

Recently uploaded

AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...
AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...
AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...
Product School
 
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
UiPathCommunity
 
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
Sri Ambati
 
DevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA ConnectDevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA Connect
Kari Kakkonen
 
Assuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyesAssuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyes
ThousandEyes
 
Search and Society: Reimagining Information Access for Radical Futures
Search and Society: Reimagining Information Access for Radical FuturesSearch and Society: Reimagining Information Access for Radical Futures
Search and Society: Reimagining Information Access for Radical Futures
Bhaskar Mitra
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance
 
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
James Anderson
 
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
Product School
 
Knowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and backKnowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and back
Elena Simperl
 
How world-class product teams are winning in the AI era by CEO and Founder, P...
How world-class product teams are winning in the AI era by CEO and Founder, P...How world-class product teams are winning in the AI era by CEO and Founder, P...
How world-class product teams are winning in the AI era by CEO and Founder, P...
Product School
 
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
Product School
 
ODC, Data Fabric and Architecture User Group
ODC, Data Fabric and Architecture User GroupODC, Data Fabric and Architecture User Group
ODC, Data Fabric and Architecture User Group
CatarinaPereira64715
 
Essentials of Automations: Optimizing FME Workflows with Parameters
Essentials of Automations: Optimizing FME Workflows with ParametersEssentials of Automations: Optimizing FME Workflows with Parameters
Essentials of Automations: Optimizing FME Workflows with Parameters
Safe Software
 
Epistemic Interaction - tuning interfaces to provide information for AI support
Epistemic Interaction - tuning interfaces to provide information for AI supportEpistemic Interaction - tuning interfaces to provide information for AI support
Epistemic Interaction - tuning interfaces to provide information for AI support
Alan Dix
 
Bits & Pixels using AI for Good.........
Bits & Pixels using AI for Good.........Bits & Pixels using AI for Good.........
Bits & Pixels using AI for Good.........
Alison B. Lowndes
 
When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...
Elena Simperl
 
"Impact of front-end architecture on development cost", Viktor Turskyi
"Impact of front-end architecture on development cost", Viktor Turskyi"Impact of front-end architecture on development cost", Viktor Turskyi
"Impact of front-end architecture on development cost", Viktor Turskyi
Fwdays
 
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered QualitySoftware Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Inflectra
 
Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...
Product School
 

Recently uploaded (20)

AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...
AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...
AI for Every Business: Unlocking Your Product's Universal Potential by VP of ...
 
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
Dev Dives: Train smarter, not harder – active learning and UiPath LLMs for do...
 
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
 
DevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA ConnectDevOps and Testing slides at DASA Connect
DevOps and Testing slides at DASA Connect
 
Assuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyesAssuring Contact Center Experiences for Your Customers With ThousandEyes
Assuring Contact Center Experiences for Your Customers With ThousandEyes
 
Search and Society: Reimagining Information Access for Radical Futures
Search and Society: Reimagining Information Access for Radical FuturesSearch and Society: Reimagining Information Access for Radical Futures
Search and Society: Reimagining Information Access for Radical Futures
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
 
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
GDG Cloud Southlake #33: Boule & Rebala: Effective AppSec in SDLC using Deplo...
 
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
 
Knowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and backKnowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and back
 
How world-class product teams are winning in the AI era by CEO and Founder, P...
How world-class product teams are winning in the AI era by CEO and Founder, P...How world-class product teams are winning in the AI era by CEO and Founder, P...
How world-class product teams are winning in the AI era by CEO and Founder, P...
 
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
De-mystifying Zero to One: Design Informed Techniques for Greenfield Innovati...
 
ODC, Data Fabric and Architecture User Group
ODC, Data Fabric and Architecture User GroupODC, Data Fabric and Architecture User Group
ODC, Data Fabric and Architecture User Group
 
Essentials of Automations: Optimizing FME Workflows with Parameters
Essentials of Automations: Optimizing FME Workflows with ParametersEssentials of Automations: Optimizing FME Workflows with Parameters
Essentials of Automations: Optimizing FME Workflows with Parameters
 
Epistemic Interaction - tuning interfaces to provide information for AI support
Epistemic Interaction - tuning interfaces to provide information for AI supportEpistemic Interaction - tuning interfaces to provide information for AI support
Epistemic Interaction - tuning interfaces to provide information for AI support
 
Bits & Pixels using AI for Good.........
Bits & Pixels using AI for Good.........Bits & Pixels using AI for Good.........
Bits & Pixels using AI for Good.........
 
When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...
 
"Impact of front-end architecture on development cost", Viktor Turskyi
"Impact of front-end architecture on development cost", Viktor Turskyi"Impact of front-end architecture on development cost", Viktor Turskyi
"Impact of front-end architecture on development cost", Viktor Turskyi
 
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered QualitySoftware Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
Software Delivery At the Speed of AI: Inflectra Invests In AI-Powered Quality
 
Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...
 

Adaptive Resonance Theory

  • 1.
  • 2. ď‚ž Unsupervised ANN ď‚ž Stability-Plasticity Dilemma ď‚ž Adaptive Resonance Theory basics ď‚ž ART Architecture ď‚ž Algorithm ď‚ž Types of ART NN ď‚ž Applications ď‚ž References
  • 3. ď‚ž Usually 2-layer ANN ď‚ž Only input data are given ď‚ž ANN must self-organise output ď‚ž Two main models: Kohonen’s SOM and Grossberg’s ART ď‚ž Clustering applications Output layer Feature layer
  • 4.
  • 5. Every learning system faces the plasticity-stability dilemma. ď‚ž The plasticity-stability dilemma poses few questions :
  • 6. ART stands for "Adaptive Resonance Theory", invented by Stephen Grossberg in 1976. ď‚žART represents a family of neural networks. ď‚žThe basic ART System is an unsupervised learning model. ď‚žThe term "resonance" refers to resonant state of a neural network in which a category prototype vector matches close enough to the current input vector. ART matching leads to this resonant state, which permits learning. The network learns only in its resonant state.
  • 7. The key innovation of ART is the use of “expectations.” › As each input is presented to the network, it is compared with the prototype vector that is most closely matches (the expectation). › If the match between the prototype and the input vector is NOT adequate, a new prototype is selected. In this way, previous learned memories (prototypes) are not eroded by new learning.
  • 8. Input Layer 1 (Retina) Layer 2 (Visual Cortex) LTM (Adaptive Weights) STM Normalization Constrast Enhancement Basic ART architecture Grossberg competitive network
  • 9. ď‚ž The L1-L2 connections are instars, which performs a clustering (or categorization) operation. When an input pattern is presented, it is multiplied (after normalization) by the L1-L2 weight matrix. ď‚ž A competition is performed at Layer 2 to determine which row of the weight matrix is closest to the input vector. That row is then moved toward the input vector. ď‚ž After learning is complete, each row of the L1-L2 weight matrix is a prototype pattern, which represents a cluster (or a category) of input vectors.
  • 10. ď‚ž Learning of ART networks also occurs in a set of feedback connections from Layer 2 to Layer 1. These connections are outstars which perform pattern recall. ď‚ž When a node in Layer 2 is activated, this reproduces a prototype pattern (the expectation) at layer 1. ď‚ž Layer 1 then performs a comparison between the expectation and the input pattern. ď‚ž When the expectation and the input pattern are NOT closely matched, the orienting subsystem causes a reset in Layer 2.
  • 11. ď‚ž The reset disables the current winning neuron, and the current expectation is removed. ď‚ž A new competition is then performed in Layer 2, while the previous winning neuron is disable. ď‚ž The new winning neuron in Layer 2 projects a new expectation to Layer 1, through the L2-L1 connections. ď‚ž This process continues until the L2-L1 expectation provides a close enough match to the input pattern.
  • 12. ď‚ž Bottom-up weights bij ď‚ž Top-down weights tij › Store class template ď‚ž Input nodes › Vigilance test › Input normalisation ď‚ž Output nodes › Forward matching ď‚ž Long-term memory › ANN weights ď‚ž Short-term memory › ANN activation pattern top down bottom up (normalised)
  • 13.
  • 14. ď‚ž The basic ART system is unsupervised learning model. It typically consists of › a comparison field and a recognition field composed of neurons, › a vigilance parameter, and › a reset module
  • 15. ď‚ž Comparison field › The comparison field takes an input vector (a one-dimensional array of values) and transfers it to its best match in the recognition field. Its best match is the single neuron whose set of weights (weight vector) most closely matches the input vector. ď‚ž Recognition field › Each recognition field neuron, outputs a negative signal proportional to that neuron's quality of match to the input vector to each of the other recognition field neurons and inhibits their output accordingly. In this way the recognition field exhibits lateral inhibition, allowing each neuron in it to represent a category to which input vectors are classified.
  • 16. ď‚ž Vigilance parameter › After the input vector is classified, a reset module compares the strength of the recognition match to a vigilance parameter. The vigilance parameter has considerable influence on the system. ď‚ž Reset Module › The reset module compares the strength of the recognition match to the vigilance parameter. › If the vigilance threshold is met, then training commences.
  • 17. Adapt winner node Initialise uncommitted node new pattern categorisation known unknown recognition comparison ď‚ž Incoming pattern matched with stored cluster templates ď‚ž If close enough to stored template joins best matching cluster, weights adapted ď‚ž If not, a new cluster is initialised with pattern as template
  • 18. ď‚ž ART-1 › Binary input vectors › Unsupervised NN that can be complemented with external changes to the vigilance parameter ď‚ž ART-2 › Real-valued input vectors
  • 19. ď‚ž ART-3 › Parallel search of compressed or distributed pattern recognition codes in a NN hierarchy. › Search process leads to the discovery of appropriate representations of a non stationary input environment. › Chemical properties of the synapse emulated in the search process
  • 20. 1 2 3 1 2 3 4Input layer Output layer with inhibitory connections ),( 3,44,3 tb The ART-1 Network
  • 21. • Mobile robot control • Facial recognition • Land cover classification • Target recognition • Medical diagnosis • Signature verification
  • 22. ď‚ž S. Rajasekaran, G.A.V. Pai, “Neural Networks, Fuzzy Logic and Genetic Algorithms”, Prentice Hall of India, Adaptive Resonance Theory, Chapter 5. ď‚ž Jacek M. Zurada, “Introduction to Artificial Neural Systems”, West Publishing Company, Matching & Self organizing maps, Chapter 7. ď‚ž Adaptive Resonance Theory, Soft computing lecture notes, “http://www.myreaders.info/html/soft_computing.html”