Calculus of Thought: Neuromorphic Logistic Regression in Cognitive Machines
Availability: Usually ships in 24 hours
Ships from and sold by elmnobooks
Calculus of Thought: Neuromorphic Logistic Regression in Cognitive Machines is a must-read for all scientists about a very simple computation method designed to simulate big-data neural processing. This book is inspired by the Calculus Ratiocinator idea of Gottfried Leibniz, which is that machine computation should be developed to simulate human cognitive processes, thus avoiding problematic subjective bias in analytic solutions to practical and scientific problems.
The reduced error logistic regression (RELR) method is proposed as such a "Calculus of Thought." This book reviews how RELR's completely automated processing may parallel important aspects of explicit and implicit learning in neural processes. It emphasizes the fact that RELR is really just a simple adjustment to already widely used logistic regression, along with RELR's new applications that go well beyond standard logistic regression in prediction and explanation. Readers will learn how RELR solves some of the most basic problems in today’s big and small data related to high dimensionality, multi-colinearity, and cognitive bias in capricious outcomes commonly involving human behavior.
"…Rice argues that cognitive machines will need to be neuromorphic, that is, based upon neuroscience, in order to simulate aspects of human cognition. He sets out the most fundamental and important concepts in modern cognitive neuroscience, including neural dynamics, implicit and explicit learning, neural synchrony, Hebbian spike-timing dependent plasticity, and neural Darwinism."--ProtoView.com, February 2014
From the Back Cover
Calculus of Thought: Neuromorphic Logistic Regression in Cognitive Machines is a must read for all scientists about a very simple computation method designed to simulate big data neural processing. This book is inspired by the Calculus Ratiocinator idea of Gottfried Leibniz which is that machine computation should be developed to simulate human cognitive processes and thus avoid problematic subjective bias in analytic solutions to practical and scientific problems. The Reduced Error Logistic Regression (RELR) method is proposed to be such a Calculus of Thought, as this book reviews how RELR’s completely automated processing may parallel important aspects of both explicit and implicit learning in neural processes. The fact that RELR is really just a simple adjustment to already widely used logistic regression is emphasized, along with RELR’s new applications that go well beyond standard logistic regression in both prediction and explanation. Particular attention is given to how RELR solves some of the most basic problems in today’s big and small data related to high dimensionality, multicollinearity and cognitive bias in capricious outcomes often involving human behavior.
Daniel M. Rice, Ph.D. is the Principal and Senior Scientist and founder of Rice Analytics in St Louis, Missouri. He is both a cognitive neuroscientist and statistician and has been practicing advanced analytic science in either medical, academic or industry settings for his entire career. Dan is a previous recipient of an Individual National Research Service Award from the National Institutes of Health and is author of more than 20 academic and industry publications in cognitive neuroscience, statistics, machine learning, and analytics. In the early 1990’s, he was the lead author on two papers that related explicit memory performance and temporal lobe brain measures to make the initial discovery and claim that Alzheimer’s disease must have an average preclinical causal period of at least 10 years. Since that time, he has worked to develop automated machine learning methods that simulate basic explicit and implicit cognitive neural processes and allow most likely solutions that avoid traditional problems related to error and bias.
About the Author
Daniel M. Rice is Principal and Senior Scientist of Rice Analytics. He founded the business in early 1996 as a sole proprietorship, but it was incorporated into its current structure in 2006. Prior to 1996, he was an assistant professor at the University of California-Irvine and the University of Southern California. Dan has almost 25 years of research project and advanced statistical modeling experience for major organizations that include the National Institute on Aging, Eli Lilly, Anheuser-Busch, Sears Portrait Studios, Hewlett-Packard, UBS, and Bank of America. He has a Ph.D. from the University of New Hampshire in Cognitive Neuroscience and Postdoctoral training in Applied Statistics from the University of California-Irvine. Dan is a previous recipient of an Individual National Research Service Award from the National Institutes of Health and is author of more than 20 publications, many of which are in conference proceedings and peer-reviewed journals in cognitive neuroscience and statistics.
Most helpful customer reviews
2 of 2 people found the following review helpful.
Good book with some provacative ideas - feels like this ...
By Amazon Customer
Good book with some provacative ideas - feels like this was a formal articulation of a doctoral dissertation. Wish there were some very practical examples so the reader could see how the new formulation worked, so it could be used. Glad I read it. My interest is in cognitive computing.
5 of 6 people found the following review helpful.
Ups the Ante on Cognitive Modeling
By Jeff Schwartz
It is difficult enough for authors to integrate diverse subject matter. Doing so while advancing the state-of-the-art is something else altogether. Daniel Rice's book Calculus of Thought achieves both goals in sparkling fashion. The reader is taken on a tour of machine learning, cognitive science, neural science, and psychometrics with a eye toward advancing the science of predictive modeilng. Just enough neural science is provided to intrigue machine learning specialists; just enough cognitive science is offered as motivation for predictive modelers to consider "explanation" on and equal footing with "prediction" when considering modeling strategy. This "just enough" integration of subject-matter extends to Rice's use of historical perspective as well.
The tour-de-force, however, is Rice's integration of psychometrics--especially consideration of measurement error--into the realm of predictive modeing. The book is worth purchasing for this reason alone, Riee convincely demonstrates the advantages of his Reduced Error Logistic Regression (RELR) modeling approach over standard approaches, and many readers will have not had exposure to this aspect of improving predictive models.
For advanced practitioners, the book offers a thoughtul analysis of a thorny issue within predictive modeling, that of attribute selection How can we predict well when we don't use the most appropriate atributes to do so? Rice argues that RELR can be used to provide a rational approach to attiribute selecition, in a way that addresses oft-ignored pschometric issues.
Calculus of Thought gives the predicitve modeling reader some badly needed ammunition:: reasons for doing predictive modeling in a certain way. The first set of reasons are psychometric, the second set of reasons are conceptual, dealing with how the human neurocognitive system addresses prediction. Imagine it-going to your boss and explaining that you are addressing both short-term and long-term aspects of your predictive model because--after all--that's what the human brain does. When your boss complains, you can explain that doing so helps move predictive modeling from craftsmanship to science--and Rice will have succeeded in his mission.
--Jeffrey P. schwartz, PhD, Cognitive Scientist