The following is a information on book which readers of this list
might find of interest.
CONSEQUENCE DRIVEN SYSTEMS
- Teaching, Learning, and Self-Learning Agents -
by Stevo Bozinovski
*201 pages
*79 figures
*27 algorithm descriptions
*8 tables
*156 references
Among its special features, the book
-------------------------------------------
** provides a unified theory of consequence driven systems
including closed-loop teaching, reinforcement learning, and
self-reinforcement learning
** describes a generic architecture of a neuro-genetic agent capable
of performing in all the mentioned paradigms, 1) consequence driven
teaching 2) external reinforcement based learning, and 3)
self-reinforcement based learning
** describes the Crossbar Adaptive Array (CAA) architecture, an
early (1981) connectionist network and explains how the CAA
architecture was the first neural network that solved a delayed
reinforcement learning task
** explains how the 1981 CAA learning method (shown on the cover of
the book) is actually the well known, 1989 rediscovered, Q-learning
method
** explains how CAA uses its genotype and phenotype (behavioral)
environment during as optimization environments in Lamarckian sense
** introduces new types of neurons, denoted as Provoking Adaptive
Units, axon provoked neurons for distributed DP tasks
** illustrates the usage of those neurons as routers in a
routing-in-networks-with-faults task.
** uses the parallel programming technique in describing the
algorithms throughout the book
*******************************************************************
Ordering information
ISBN 9989-684-06-5, Gocmar Press
price: USD $15, paperback
send email to the publisher representative:
Ivan PopStefanija, Gocmar Press, USA
ivan at alex.ecs.umass.edu
********************************************************************
CONTENTS:
1. INTRODUCTION
1.1. The framework
1.2. Agents and architectures
1.3. Neural architectures
1.3.1 Greedy policy neural architectures
1.3.2. Recurrent architectures
1.3.3. Crossbar architectures
1.3.4. Subsumption architecture adaptive arrays
1.4. Problems. Emotional Graphs
1.5. games. Emotional petri nets
1.6. Parallel programming
1.7. Bibliographical and other notes
2. CONSEQUENCE LEARNING AGENTS
2.1. Thhe agent-environment interface
2.2. A taxonomy of learning paradigms
2.3. Classes of consequence learning agents
2.4. A generic consequence learning architecture
2.5. Learning rules and routines
2.6. Bibliographical and other notes
3. CONSEQUENCE DRIVEN TEACHING
3.1. Class T agents
3.2. Learners
3.2.1. Multi layer perceptrons
3.2.2. Greedy policy neural arrays
3.3. Teachers
3.3.1. Toward a theory of teaching systems
3.3.2. Teaching strategies
3.4. Curriculums
3.4.1. Curriculum grammars amd languages
3.4.2. Curriculum space approach
3.5. Pattern classification teaching as integer programming
3.6. Pattern classification teaching as Dynamic Programming
3.7. Bibliographical and other notes
4. EXTERNAL REINFORCEMENT LEARNING
4.1. Reinforcement learning NG agents
4.2. Associative Serach Network (ASN)
4.2.1. Basic ASN
4.2.2. Reinforcement predictive ASN
4.3. Actor-Critic architecture
4.4. Bibliographical and other notes
5. SELF-REINFORCEMENT LEARNING
5.1. Conceptual framework
5.2. Self-reinforcement learning and the NG agents
5.3. The Crossbar Adaptive Array architecture
5.4. How it works
5.4.1. Defining primary goals from thew genetic environment
5.4.2. Secondary reinforcement mechanism
5.4.3. The CAA learning method
5.5. Example of a CAA architecture
5.6. Solving problems with a CAA architecture
5.6.1. Learning in emotional graphs: maze running
5.6.2. Learning in loosely defined emotional graphs: Pole balancing
5.7. Another example of a CAA architecture
5.8. Using entropy in Markov decision Processes
5.9. Issues on the genetic environment
5.9.1. CAA as an optimization architecture
5.9.2. Complelemtarity with the Genetic Algorithms
5.9.3. Self-reinforcement: genetic environment approach
5.10. Bibliographical and other notes
6. CONSEQUENCE PROGRAMMING
6.1. Dynamic Programming and markov decision Problems
6.2. Inroducing cost in the CAA architecture
6.3. Q-learning
6.4. A taxonomy of the CAA-method based learning algorithms
6.5. Producing optimal solution in a stochastic environment
6.6. Distributed Consequence Programming: A neural theory
6.6.1. Provoking units: axon provoked neurons
6.6.2. An illustration: Routing in client-server networks with faults
6.7. Bibliographical nad other notes
7. SUMMARY
8. REFERENCES
9. INDEX
-----------------------------------------------------------------------
********************************************
Ordering Information
Book: CONSEQUENCE DRIVEN SYSTEMS
Author: Stevo Bozinovski
ISBN: 9989-684-06-5
Publisher: GOCMAR Press, 1995
Price: $15
Shipping & handling
Domestic:
Regular mail add $ 2.50
Priority mail add $ 4.00
COD add $ 8.00
International:
Air mail add $ 8.00
Make check payable to: GOCMAR Press USA
Mail your order with the payment to:
GOCMAR Press USA
c/o Ivan PopStefanija
15 Whippletree ln
Amherst, MA 01002-3153
********************************************
Thank you
Ivan PopStefanija
e-mail: <ivan at alex.ecs.umass.edu>
GOCMAR Press USA
