Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Growing adaptive machines : combining development and learning in artificial neural networks
Kowaliw T., Bredeche N., Doursat R., Springer Publishing Company, Incorporated, New York, NY, 2014. 261 pp. Type: Book (978-3-642553-36-3)
Date Reviewed: Mar 25 2015

Since I first began researching neural networks 25 years ago, I have been wondering how to make them biologically more plausible. Thus, I am pleased that this book considers the importance of biological plausibility in artificial neural networks (ANNs). Respected researchers in the field wrote the chapters. The brilliant first chapter, written by the editors, provides the fundamentals of artificial neurogenesis and a discussion of what the remaining chapters contain. Important concepts are presented and discussed, including deep learning (DL), which is a popular topic these days in machine learning. It also presents a list of more than 300 essential references.

Chapter 2 addresses probabilistic learning and neuroscience. The author presents several kinds of learning, aiming to link it to biology. Chapter 3 was written by the father of DL and, in addition to discussing DL, focuses on evolving culture. According to the author, “cultural evolution occurs on a much faster scale than genetic evolution.” He proposes a series of hypotheses in order to make the ideas clearer, and in this the author was successful. He also discusses the memes that should be transmitted between minds. In this case, novelty is awarded: novel memes that are further away from existing ones yield more diversity in the solutions explored.

Chapter 4 discusses auto-associators as building blocks for deep neural networks. The authors propose an architecture called a sparse auto-associator that uses stacks in deep learning. The authors also present perspectives on future research, namely deep learning combined with auto-associators and convolution. Chapter 5 presents a review of hypercube-based neuroevolution of augmenting topologies (HyperNEAT) applications in its first five years. Chapter 6 discusses genetic regulatory evolving artificial networks (GReaNs). Chapter 7 introduces a visual model that could have important applications outside of neuroscience. The chapter also highlights the importance of knowing neuroscience jargon in order to understand biologically plausible ANN architectures. Chapter 8 presents two approaches to the evolution and development of ANN, neurocentric and holocentric, both of which are related to learning. The author states that the neuron is extremely complex, so how do we create a computational model from it? He argues that holocentric ANN is more efficient than neurocentric, but is not as biologically plausible. Finally, chapter 9 closes the book with a presentation of artificial evolution and plasticity concepts.

Although the book has some typographical errors and (unfortunately, as usual) black-and-white figures are treated as if they are in color, the book is recommended for those who want to know more about ANNs and their biologically inspired architectures, especially those related to learning.

Reviewer:  João Luís G. Rosa Review #: CR143278 (1506-0459)
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
Learning (I.2.6 )
 
 
Self-Modifying Machines (F.1.1 ... )
 
 
Applications And Expert Systems (I.2.1 )
 
 
Models Of Computation (F.1.1 )
 
Would you recommend this review?
yes
no
Other reviews under "Learning": Date
Learning in parallel networks: simulating learning in a probabilistic system
Hinton G. (ed) BYTE 10(4): 265-273, 1985. Type: Article
Nov 1 1985
Macro-operators: a weak method for learning
Korf R. Artificial Intelligence 26(1): 35-77, 1985. Type: Article
Feb 1 1986
Inferring (mal) rules from pupils’ protocols
Sleeman D.  Progress in artificial intelligence (, Orsay, France,391985. Type: Proceedings
Dec 1 1985
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy