Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

Carleton University

$25,000.00

Learning, neural and pattern recognition systems

FY 2012-13

Summary
During the term of the award, I propose to undertake various projects which concern both the theory and applications of Learning Automata (LA), and Pattern Recognition (PR) as listed below: 1. I propose to develop a formal LA-based theory for Tutorial-like systems in which an imperfect ""Teacher"" teaches from inaccurate teaching material, and where each individual probabilistic ""Student"" found in a ""Classroom"" of such ""Students"" can learn from the ""Teacher"" and his ""Colleagues"", where each of these entities need not be a human being but for example, a component in a cybernetic system. 2. The families of Pursuit and Estimator-based LA have been characterized by using LA which use action probabilities, but which also maintain estimates of the reward probabilities. In the past, these estimates have been of the Maximum Likelihood nature. I propose to develop the theory of the Bayesian Estimator LA, in which the estimates used to design the Estimator LA are obtained in a Bayesian manner. 3. The Goore Game (GG) is a non-trivial non-zero-sum game, which unlike the games traditionally studied in the AI literature (e.g., Chess, Checkers, Lights-Out etc.), is essentially a distributed game. LA have been used to play the game when all the parameters are unknown. I propose to use LA to solve the GG when the ""players"" can also be traitors, and to study its applications in sensor networks and distributed optimization. 4. I propose to use LA to solve the COMMONS game, a generalization of the Prisoner's Dilemma. This game simulates a society in which Entities have to work cooperatively so as to utilize common shared resources. 5. I propose to use LA to enhance the Growing Self-Organizing Map (GSOM) which is a neural network clustering algorithm based on Kohonen's Self-Organizing Map. 6. Historically, the concept of designing Neural Networks (NN) which are inherently chaotic and which also possess PR capabilities by abruptly switching to periodicity, is new. Our prior research has led to the only reported NNs that demonstrate Chaotic PR phenomena. I propose to study various problems in this area.
Program
Discovery Grants Program - Individual
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
2489-2010
Name nom
Oommen, John
Department département
Computer Science, School of
Organization id
23
Institution établissement
Carleton University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2012
Competition year année de concours
2010
Award amount
25000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2606
Research subject group en
Robotics
Research subject group fr
Robotique
Research subject en
Systems, man and cybernetics
Research subject fr
Systèmes, homme et cybernétique
Partie
3
Nb partie
5
Application title
Learning, neural and pattern recognition systems
Keyword
NSERC
Application summary
During the term of the award, I propose to undertake various projects which concern both the theory and applications of Learning Automata (LA), and Pattern Recognition (PR) as listed below: 1. I propose to develop a formal LA-based theory for Tutorial-like systems in which an imperfect ""Teacher"" teaches from inaccurate teaching material, and where each individual probabilistic ""Student"" found in a ""Classroom"" of such ""Students"" can learn from the ""Teacher"" and his ""Colleagues"", where each of these entities need not be a human being but for example, a component in a cybernetic system. 2. The families of Pursuit and Estimator-based LA have been characterized by using LA which use action probabilities, but which also maintain estimates of the reward probabilities. In the past, these estimates have been of the Maximum Likelihood nature. I propose to develop the theory of the Bayesian Estimator LA, in which the estimates used to design the Estimator LA are obtained in a Bayesian manner. 3. The Goore Game (GG) is a non-trivial non-zero-sum game, which unlike the games traditionally studied in the AI literature (e.g., Chess, Checkers, Lights-Out etc.), is essentially a distributed game. LA have been used to play the game when all the parameters are unknown. I propose to use LA to solve the GG when the ""players"" can also be traitors, and to study its applications in sensor networks and distributed optimization. 4. I propose to use LA to solve the COMMONS game, a generalization of the Prisoner's Dilemma. This game simulates a society in which Entities have to work cooperatively so as to utilize common shared resources. 5. I propose to use LA to enhance the Growing Self-Organizing Map (GSOM) which is a neural network clustering algorithm based on Kohonen's Self-Organizing Map. 6. Historically, the concept of designing Neural Networks (NN) which are inherently chaotic and which also possess PR capabilities by abruptly switching to periodicity, is new. Our prior research has led to the only reported NNs that demonstrate Chaotic PR phenomena. I propose to study various problems in this area.