Asim Roy

Asim Roy, Ph.D. is a Professor of Information Systems at Arizona State University. He received his B.E. in Mechanical Engineering from Calcutta University, India, his M.S. in Operations Research from Case Western Reserve University, Cleveland, Ohio, and his Ph.D. in Operations Research from the University of Texas at Austin. He has been a Visiting Scholar at Stanford University, visiting Professor David Rumelhart in the Psychology Department, and a Visiting Scientist at the Oak Ridge National Laboratory, Tennessee. 

Designed and developed IFPS/OPTIMUM
Asim designed and developed the software system IFPS/OPTIMUM in the 1980s that pioneered the idea of incorporating optimization tools in financial and planning languages for managerial use. It has been used by hundreds of corporations worldwide for financial, corporate, and production planning. The system had saved many companies hundreds of millions of dollars. Following in its footsteps, such optimization systems are now widely available with spreadsheet systems such as Excel Solver within Excel and extensively used in graduate schools to teach modeling for optimization.

Explainable AI (XAI) for Computer Vision/Object Recognition - A Paradigm Shift 

Asim has developed a new approach to Explainable AI (XAI) for computer vision and object recognition. In current approaches, we generally build models first and then try to explain its predictions by various means such as saliency maps and gradcams. In this new approach, the form of explanation is specified first and then a model is created according to those specifications. The process is very similar to where an architect meets with clients first to determine their needs before designing a house. 

Asim’s brain theories
Asim has published four theories of the brain. The first theory postulates that localist representation, as opposed to distributed representation, is widely used in the brain: A theory of the brain: localist representation is used widely in the brain. That implies that firings of neurons in the brain have “meaning and interpretation” on a stand-alone basis. The second theory postulates that grandmother cells are used widely in the brain: An extension of the localist representation theory: grandmother cells are also widely used in the brain. Grandmother cells are a special type of localist cells and represent complex concepts that are multimodal invariant. A third theory postulates that a purely abstract cognitive system exists in the brain at different levels of processing: The Theory of Localist Representation and of a Purely Abstract Cognitive System: The Evidence from Cortical Columns, Category Cells, and Multisensory Neurons. In 2008, Asim published a theory of the brain that postulates that there are parts of the brain that control other parts and thus control theoretic principles can be used to design and construct systems like the brain: Connectionism, controllers and a brain theory. These four theories invalidate many ideas of the current dominant theory of the brain called “Connectionism.” 

Phys.org reported on Asim’s brain theories

Phys.org did the following reports on Asim’s brain theories:

1.      On grandmother cells: If you can't beat them, join them: Grandmother cells revisited

2.      On localist representation: Do brain cells need to be connected to have meaning?

3.      On the controller theory of the brain: Professor Finally Publishes Controversial Brain Theory 

James McClelland of Stanford University and David Plaut of Carnegie Mellon University added their criticism of the localist representation theory in this Phys.org report: Do brain cells need to be connected to have meaning?  Here’s the response to their criticism: Response to Plaut and McClelland in the Phys.org story.  

Neuroscience pioneer Horace Barlow defends grandmother cell theory

The grandmother cell theory was strongly supported by late Prof. Horace Barlow of Cambridge University (Horace Barlow - Wikipedia), a neuroscience pioneer and great-grandson of Charles Darwin. In his later years, he became a “single unit freak.” About grandmother cells, he writes in a personal note to me in 2012, “though I still hold that something conforming amazingly well to what was conceptualised by Jerry Lettvin 50 years ago, really does exist!... and of course the more widely discussed it is the better pleased I shall be, though I fear that what I have written will not be universally accepted, at least at first!” The value of the single cell theory is that it gets us to abstract concepts and symbolic reasoning beyond pattern recognition of connectionist systems. And we have plenty of neurophysiological evidence for single cell abstractions in the brain.

Professional Activities

Asim was on the Governing Board of the International Neural Network Society (INNS) from 2012 to 2021. He is the founder of three INNS Sections: (1) Autonomous Machine Learning (AML Section), (2) Explainable AI (XAI Section (inns.org)) and (3) AI, Law, and Regulation (AI, Law, and Regulation Section). He also started the Virtual Technical Event series of INNS (Virtual Technical Events (inns.org)).

He was the Guest Editor-in-Chief of an open access eBook Representation in the Brain of Frontiers in Psychology. He was also the Guest Editor-in-Chief of two special issues of Neural Networks | Journal | ScienceDirect.com by Elsevier – one on autonomous learning and the other on big data analytics. He was also the Guest Editor-in-Chief of the special issue of Cognitive Computation on What AI and Neuroscience Can Learn from Each Other—Open Problems in Models and Theories. He is the Senior Editor of the Big Data Analytics section of Cognitive Computation | Home (springer.com) and served on the editorial boards of other journals.

He has served on the organizing committees of many scientific conferences. He started the Big Data conference series of INNS and was the General Co-Chair of the first one in San Francisco in 2015. He was the Technical Program Co-Chair of IJCNN 2015 (International Joint Conference on Neural Networks) in Ireland and the IJCNN Technical Program Co-Chair for the World Congress on Computational Intelligence 2018 (WCCI2018) in Rio de Janeiro, Brazil. He was the IJCNN Conference Chair for WCCI 2020 in Glasgow, UK. He was the Program Chair for the ORSA/TIMS (Operations Research Society of America / The Institute of Management Sciences) National meeting in Las Vegas in 1990 and General Chair of the ORSA/TIMS National meeting in Phoenix in 1993. 

He has been listed in Who’s Who in America.