My long-term scientific goal is to try to work out how the brain learns (self-organises). This took me in directions of Information Theory and probability theory for neural networks. This provides a hopelessly crude and impoverished model (called redundancy reduction) of what the brain does and how it lives in its world. Unfortunately, it's the best we have at the moment. We have to do some new mathematics before we reach self-organisational principles that will apply to the physical substrate of the brain, which is molecular: ion channels, enzyme complexes, gene expression networks. We have to think about dynamics, loops, open systems, how open dynamical systems can encode and effect the spatio-temporal trajectories of their perturbing inputs.
I just wrote a paper about all this if you are interested. It was an invited contribution to the December 1999 special Millennial issue of the Philosophical Transactions of the Royal Society of London (apparently the world's oldest journal), Biological Sciences section. We were supposed to predict the future of our fields and the effects on society. Here it is in PDF form [with '(', the right bracket, strangely absent from the text]:
Levels and Loops:
the future of Artificial Intelligence and Neuroscience
(the reference is Phil. Trans. R. Soc. Lond. B (1999) 354, 2013-2020)
In retrospect, it is pretty negatively written, but I think there are
some good points.
Send me email if you want to argue...
Maybe you are still interested in Infomax/ICA/blind separation and all
that? (OK, I am too.)
If so **HERE IS MY ICA WEB PAGE.** It has many papers, some code, and other ICA links.
My ftp site is here.
Check out the README file for a description of what's what there.
***********New Contact Co-ordinates.
There is no mental ecologist of the week this week.
11/1/2000 -Tony Bell