Comments on: Jeff Hawkins, Neurodudes-style http://neurodudes.com/2005/03/24/jeff-hawkins-neurodudes-style/ at the intersection of neuroscience and AI. Fri, 29 Aug 2008 19:38:59 +0000 http://wordpress.org/?v=2.6 By: neurodudes » Blog Archive » Company Using “In Silico Embodiment” To Build Artificial Intelligence http://neurodudes.com/2005/03/24/jeff-hawkins-neurodudes-style/#comment-325166 neurodudes » Blog Archive » Company Using “In Silico Embodiment” To Build Artificial Intelligence Fri, 24 Aug 2007 05:02:08 +0000 http://s93794016.onlinehome.us/wordpress/?p=106#comment-325166 [...] Goertzel’s presentation is notable for its modesty. He acknowledges the classic problems of the AI enterprise, like the habit of promising AI algorithms failure to scale up due to combinatorial explosion, and of AI presentations, like hiding inconvenient details behind complex Powerpoint presentations. Refreshingly, he makes fun of these bad patterns as cliches to reassure the audience that he wants to avoid falling into these traps. It pays off, and he wins my respect for doing so. He lays out some of the details of his architecture, but my attention is drawn more to his basic philosophy and approach more than anything else. He makes some interesting points about combinatorial complexity of AI programs, and, in the QA period, offers his views on competing architectures like that of Jeff Hawkins at Numenta. [...] [...] Goertzel’s presentation is notable for its modesty. He acknowledges the classic problems of the AI enterprise, like the habit of promising AI algorithms failure to scale up due to combinatorial explosion, and of AI presentations, like hiding inconvenient details behind complex Powerpoint presentations. Refreshingly, he makes fun of these bad patterns as cliches to reassure the audience that he wants to avoid falling into these traps. It pays off, and he wins my respect for doing so. He lays out some of the details of his architecture, but my attention is drawn more to his basic philosophy and approach more than anything else. He makes some interesting points about combinatorial complexity of AI programs, and, in the QA period, offers his views on competing architectures like that of Jeff Hawkins at Numenta. [...]

]]> By: Norman Vander http://neurodudes.com/2005/03/24/jeff-hawkins-neurodudes-style/#comment-208345 Norman Vander Mon, 02 Jul 2007 20:01:48 +0000 http://s93794016.onlinehome.us/wordpress/?p=106#comment-208345 […]I just read “On Intelligence”, and I was really impressed. Numenta isn’t just about a new platform, API, algorithm.[…] […]I just read “On Intelligence”, and I was really impressed. Numenta isn’t just about a new platform, API, algorithm.[…]

]]> By: neurodudes » Blog Archive » Hawkins Releases Numenta Code http://neurodudes.com/2005/03/24/jeff-hawkins-neurodudes-style/#comment-47611 neurodudes » Blog Archive » Hawkins Releases Numenta Code Wed, 07 Mar 2007 04:12:46 +0000 http://s93794016.onlinehome.us/wordpress/?p=106#comment-47611 [...] We have been covering Hawkins’ work for a while now. See these previous posts for more background info. [...] [...] We have been covering Hawkins’ work for a while now. See these previous posts for more background info. [...]

]]> By: Neville http://neurodudes.com/2005/03/24/jeff-hawkins-neurodudes-style/#comment-656 Neville Tue, 31 May 2005 22:59:41 +0000 http://s93794016.onlinehome.us/wordpress/?p=106#comment-656 8 hours! That's not my idea of an easy read... must have been a real page-turner! Seriously though, thanks for the mini book review. I'll have to check it out. 8 hours! That’s not my idea of an easy read… must have been a real page-turner!

Seriously though, thanks for the mini book review. I’ll have to check it out.

]]> By: Tevye R. Krynski http://neurodudes.com/2005/03/24/jeff-hawkins-neurodudes-style/#comment-654 Tevye R. Krynski Wed, 31 Dec 1969 19:00:00 +0000 http://s93794016.onlinehome.us/wordpress/?p=106#comment-654 I just read "On Intelligence", and I was really impressed. Numenta isn't just about a new platform, API, algorithm. It's not even about a new kind of neural network. It's about Hawkins' theory of the cortex, and it is truly revolutionary (whether it's accurate or not). The idea is that every region of the cortex -- visual, auditory, motor, association, language -- does the same thing: detection of patterns, representation of invariances, and prediction from a generative model of the world based on invariances. He has a theory of how this is accomplished in each individual cortical column using the cells in the 6 layers. And he has a theory of how a hierarchical connectivity between cortical regions enables cognition to become more and more abstract and analogical in higher cortical regions. I highly recommend reading the book before forming any opinions. It's an easy read -- I finished it in 8 hours, and I'm a slow reader. I just read “On Intelligence”, and I was really impressed. Numenta isn’t just about a new platform, API, algorithm. It’s not even about a new kind of neural network. It’s about Hawkins’ theory of the cortex, and it is truly revolutionary (whether it’s accurate or not). The idea is that every region of the cortex — visual, auditory, motor, association, language — does the same thing: detection of patterns, representation of invariances, and prediction from a generative model of the world based on invariances. He has a theory of how this is accomplished in each individual cortical column using the cells in the 6 layers. And he has a theory of how a hierarchical connectivity between cortical regions enables cognition to become more and more abstract and analogical in higher cortical regions. I highly recommend reading the book before forming any opinions. It’s an easy read — I finished it in 8 hours, and I’m a slow reader.

]]>