Dharmendra S. Modha

My Work and Thoughts.

Archives for 2008

Horst Simon

By

Last week, at 2008 Almaden Institute, I had the privillege of inviting and introducing one of the key note speakers, Dr. Horst Simon. He gave a truly fascinating talk. Here is my introduction:

Our distinguished key note speaker today is Dr. Horst Simon. Dr. Simon holds a PhD in Mathematics from UC Berkeley.

He is best known for his breakthrough work on highly parrallel recursice spectral bisection algorithm and for his work on partitioning sparse matrices. He has won the Gordon Bell Prize and has received H. Julian Allen Award from NASA.

He has been a Senior Manager at Silicon Graphics, Computer Sciences Corporation, and Boeing, and has been on the faculty at State University of New York. From 1996-2006, he served as Director of Department of Energy’s National Energy Research Scientific Computing Center which is the flagship scientific computing facility for the Office of Science in the U.S. Department of Energy. He is currently Associate Lab director at Lawrence Berkeley National Lab and Director of Computational Research Division. He is an Adjunct Professor at UC Berkeley, and recently led creation of a multidisciplinary new initiative Computational Science and Engineering.

He has served on boards of several organizations and on editorial boards of several journals in high-performance computing arena.

In my view, he is Mr. Supercomputing in the academic world today. He is co-author of TOP 500 which is a twice-yerarly revised list of the world’s most powerful computer systems, and provides the most direct window into future of computing.

SyNAPSE: Systems of Neuromorphic Adaptive Plastic Scalable Electronics

By

DARPA’s Defense Sciences Office (DSO) has recently issued a Broad Agency Announcement entitled SyNAPSE. The program is led by Dr. Todd Hylton.

Please see here for the BAA.

Here is a brief description:

"Over six decades, modern electronics has evolved through a series of major developments (e.g., transistors, integrated circuits, memories, microprocessors) leading to the programmable electronic machines that are ubiquitous today.  Owing both to limitations in hardware and architecture, these machines are of limited utility in complex, real-world environments, which demand an intelligence that has not yet been captured in an algorithmic-computational paradigm. As compared to biological systems for example, today’s programmable machines are less efficient by a factor of one million to one billion in complex, real-world environments.  The SyNAPSE program seeks to break the programmable machine paradigm and define a new path forward for creating useful, intelligent machines. 

The vision for the anticipated DARPA SyNAPSE program is the enabling of electronic neuromorphic machine technology that is scalable to biological levels.  Programmable machines are limited not only by their computational capacity, but also by an architecture requiring (human-derived) algorithms to both describe and process information from their environment.  In contrast, biological neural systems (e.g., brains) autonomously process information in complex environments by automatically learning relevant and probabilistically stable features and associations.  Since real world systems are always many body problems with infinite combinatorial complexity, neuromorphic electronic machines would be preferable in a host of applications—but useful and practical implementations do not yet exist. 

The key to achieving the vision of the SyNAPSE program will be an unprecedented multidisciplinary approach that can coordinate aggressive technology development activities in the following areas:  1) hardware; 2) architecture; 3) simulation; and 4) environment." 

Decade of the Mind III: Emergence of Mind

By

The Decade of the Mind III event will be held at Great Apes Trust in Iowa.

Once again, the list of speakers is very distinguished, please see bios here:

Dr. Giulio Tononi
Dr. James L. Olds
Dr. Roger K. R. Thompson
Dr. Colin Allen
Dr. Kathy Schick
Dr. Nicholas Toth
Dr. Anne Russon
Dr. Tetsuro Matsuzawa
Dr. Robert Seyfarth
Dr. Merlin Donald

The reader may find the Decade of the Mind letter that we authored in Science interesting.

Kwabena Boahen: Neurogrid–Emulating a million neurons in the cortex

By

Today, I had a tremendous good fortune to host Professor Kwabena Boahen from Stanford University for a widely attended colloquium talk at Almaden. Professor Boahen is a brilliant scientist and bioengineer who seeks to emulate the brain in hardware. Professor Boahen is a protégé of Professor Carver Mead from CalTech.

My favorite papers amongst his many wonderful articles are Neuronal Ion-Channel Dynamics in Silicon and the now classic Point-to-Point Connectivity Between Neuromorphic Chips using Address-Events. 

ABSTRACT

The digital technique used to simulate neural activity has not changed since Hodgkin and Huxley pioneered ion-channel modeling in the 1950s. Since then, progress has come incrementally, from computer performance doubling every eighteen months (Moore’s Law), plateauing in recent years, and putting real-time cortex-scale simulations outside the realm of the fastest supercomputers for the foreseeable future. With recent advances in neural recording and imaging techniques, our ability to characterize the brain’s structure and function truly trumps our ability to simulate its behavior. Fortuitously, the analog technique developed by neuromorphic engineers over the past two decades has now matured, with the recently developed ability to program various types of ion-channels as well as arbitrary patterns of synaptic connections.

Exploiting the analog technique, Neurogrid will help neuroscientists vet various hypotheses by performing simulations large enough to include interactions between multiple cortical areas yet detailed enough to account for what is known about brain function and neuronal structure. While neuronal-level mechanisms have been linked to network-level functions through computational modeling (e.g., generation of brain rhythms), scaling these models up to the area- and system-levels (where cognition emerges) has proved difficult. In the visual system alone, there are three dozen cortical areas, each with its own representation of the visual scene. It is not understood how conflicting information in these areas is reconciled.

When it is completed this year, Neurogrid will emulate a million neurons in the cortex (i.e., simulate in real-time)—rivaling the performance of 20–200 IBM Blue Gene racks on this particular task—at under a thousandth the cost.

BIOGRAPHY

Professor Kwabena Boahen joined Stanford’s Bioengineering Department as Associate Professor in December 2005. From 1997 to 2005 he was on the faculty of University of Pennsylvania, Philadelphia PA. He is a bioengineer who is using silicon integrated circuits to emulate the way neurons compute, linking the seemingly disparate fields of electronics and computer science with neurobiology and medicine. His interest in neural networks developed soon after he left his native Ghana to pursue undergraduate studies in Electrical and Computer Engineering at Johns Hopkins University, Baltimore, in 1985. He went on to earn a doctorate in Computation and Neural Systems at the California Institute of Technology in 1997. His lab is currently developing Neurogrid, a specialized hardware platform that will enable the cortex’s inner workings to be simulated in detail—something outside the realm of even the fastest supercomputers. Professor Boahen’s numerous contributions to the field of neuromorphic engineering include a silicon retina that could be used to give the blind sight and a self-organizing chip that emulates the way the juvenile brain wires itself up. His scholarship is widely recognized, with over sixty publications to his name, including a cover story in the May 2005 issue of Scientific American. He has received several distinguished honors, including a Fellowship from the Packard Foundation in 1999, a CAREER award from the National Science Foundation in 2001, a Young Investigator Award from the Office of Naval Research in 2002, and the National Institute of Health Director’s Pioneer Award in 2006. The professor is an avid cyclist.