[gregdetre]

Building on Hofstadter's models of analogy-making to build pattern recognition algorithms that work in a more human-like way. At least, that's my hope!

My background is in computational neuroscience, but I'm doing this on my own, mostly for fun. If it sounds interesting, I'd love some company! greg at gregdetre dot co dot uk

[nileshtrivedi]

This sounds interesting. What kind of data structure do you use to represent a "pattern"?

[mindcrime]

I'm no AI expert, but after reading both Hawkins' On Intelligence and Kurweil's How To Create A Mind, I'm pretty sold on the idea that our minds are basically pattern-matching machines. Lately I've been trying to spend a lot of time "thinking about thinking" and recognizing how I form thoughts, relate to memories, etc. I am not super familiar with Hofstadter's ideas, since I haven't (yet) finished even GEB, but the general idea is very intriguing to me. I'd be happy to chat with you sometime and bounce ideas around. You're almost certainly way ahead of me in terms of technical knowledge in this domain though.

[gregdetre]

Cool! Drop me a line at greg at gregdetre dot co dot uk

Fluid Concepts & Creative Analogies ch 4-6 are the best introduction to his modeling work. I can't recommend them highly enough.

[mindcrime]

Excellent, I'll look into getting a copy of that. I have several of his other books (GEB, I Am A Strange Loop, Metamagical Themas, etc.) but only got about 1/3rd of the way through GEB before setting it aside to focus on other stuff for "a while". Unfortunately "a while" has turned into over a year now, so I guess it's time to get back to it. :-)