Understanding Speech

Robert Hecht-Nielsen
Adjunct Professor, Electrical & Computer Engineering




An authority on neural networks, he introduced the first comprehensive theory of the mammalian cerebral cortex and thalamus in 2002. His research revolves around scientific testing, elaboration, and extension of this theory.

Professor Hecht-Nielsen is an expert on brain theory, associative memory neural networks and Perceptron theory. His theory of thalamocortex is currently being promulgated and integrated into research worldwide.

Capsule Bio:
Robert Hecht-Nielsen has been adjunct professor at UCSD since 1986. He teaches the popular ECE 270 three-quarter graduate course Neurocomputing, which focuses on the basic constructs of his theory of thalamocortex and their applications. He is a member of the UCSD Institute for Neural Computation and is a founder of the UCSD Graduate Program in Computational Neurobiology. An IEEE Fellow, he has received the IEEE Neural Networks Pioneer Award and the ECE Graduate Teaching Award. He received his Ph.D. in Mathematics from Arizona State University in 1974.
http://www.jacobsschool.ucsd.edu/faculty/faculty_bios/findprofile.pl?fmp_reci...

Category:

Science & Technology

Tags:

• CITRIS
• UCBerkeley
• Cognitive
• Computing
• Research
• Science
• Technology

[Aoitetsugakusha]

@rolandomartin We do plenty of simulations; the lab is still young and we have just a few robots at the moment. The size of the lab is expanding pretty rapidly over the last few months, though I'll be going to a different campus for grad school in the Fall.

You should find it pretty easily if you search 'Cognitive Anteater Robotics Lab' on google. Note: The anteater is the campus mascot, hence the name.

[rolandomartin]

@Aoitetsugakusha Do you carry on computer simulations? Can you elaborate on the robos of your lab?

[Aoitetsugakusha]

@rolandomartin Well, in truth IBM's Watson is one of the best natural language processors we have, I just don't think it has anything to do with neural networks. So that may not be too far off.

Modha's model was 'to scale' with a cat brain; that doesn't mean it was made to simulate cat behavior though. This is probably what Markram is relying on.

My own lab is working with artificial rat brains, but even our best robot has only the neuromodulatory systems, which is huge but not everything.

[rolandomartin]

@Aoitetsugakusha Moreover, his claims suggest that by today (four years after this presentation of the theory) we should have fully fledged speech recognizers working on NLP and understanding, aiming to the customer service market (good luck to all the jobs in India) but nothing has been announced or presented.

I am dissapointed. Darmendra Modha claimed he simulated a cat brain. Markram wrote a letter to IBM's CTO claiming the guy was a fraud.

[rolandomartin]

I agree the progress in this matter is in the order of decades, following Kurzweil and Markram we will have a full working model of the brain in 20 or 10 years; however, what I meant on the comment, is the fact that the Confabulation theory does not explain too much. On one hand, Nielsen claims this is the ultimate theory that explains how cognition works. He shows very interesting outcomes with the simulations but the same results can be obtained with N-order Markov Chains.

[Aoitetsugakusha]

@rolandomartin It's difficult enough getting a model of a single brain region or network to be the 'predominant' model of that region or network. Hecht-Neilsen's model is what you'd call a 'global brain theory,' and those are much harder to catch on. I don't think any such global brain theory has ever reached anywhere near universally accepted status. Coupled with the comparatively few people doing computational neuroscience, that's why you haven't seen much new on it. Come back in ten years?

[TheReasonWhyGuy]

"Muscles of thought"

That's a weird way to put it...

Some people may confuse that with LITERAL muscles :P

[rolandomartin]

I must say I bought his "Confabulation Theory" book, and was quickly lured by the outrageous claims he makes about confabulation being the ultimate explanation on how the cortex works. It's been a few years since the book and I was not able to find not even a single reference about his work since then. By now I would expect he had came up with more simulations and demonstrations prooving his claims on the theory.