Re: numenta
On 02 Jun 2014, at 19:37, Telmo Menezes wrote: On Mon, Jun 2, 2014 at 7:24 PM, meekerdb meeke...@verizon.net wrote: http://numenta.org/ An organization formed by Jeff Hawkins (inventor of the Palm Pilot) to study artificial intelligence. Hawkins idea is that lower level modules in the brain continually try to predict what signals they will next receive; and it is only when the predictions fail that signals are passed up to higher (more interconnected) modules, and it is at the highest level they become conscious thoughts. I read his book On Intelligence a few years ago and recommend it. It is quite interesting and has some nice ideas on how to implement an AGI. It conveys a lot of information on neuroscience with a strong focus on the visual cortex. He even derives a machine learning model from his ideas, and it appears to have practical applications: http://en.wikipedia.org/wiki/Hierarchical_temporal_memory On the other hand, the excitement appears to have fizzled out after some initial hype, but maybe I'm just unaware of further progress. Will look into it. Pro: I really appreciate his it's time for computer science to tackle the brain attitude, focused on actually building things; Con: He dismisses the mind-body problem by essentially claiming that consciousness doesn't even exist. It annoys me, but I can tolerate it because he delivers interesting ideas and models on the practical side of things. Not only I am unable to go to that page, but my browser get really mad after I tried. Actually, I have lost my connection with the net. I guess my OS is too old. But from what you say, we might agree on this. That theory is close to the idea that consciousness is called for when automated part of the brain don't fit with the situation. We can open everyday our door without conscious thinking, but if the key appears to not function well, we get conscious of the situation. Then, if he claims that consciousness does not exist, well, we have a problem. Agreed. Bruno Best, Telmo. Brent -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. http://iridia.ulb.ac.be/~marchal/ -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
Re: numenta
On Tue, Jun 3, 2014 at 4:17 PM, Bruno Marchal marc...@ulb.ac.be wrote: On 02 Jun 2014, at 19:37, Telmo Menezes wrote: On Mon, Jun 2, 2014 at 7:24 PM, meekerdb meeke...@verizon.net wrote: http://numenta.org/ An organization formed by Jeff Hawkins (inventor of the Palm Pilot) to study artificial intelligence. Hawkins idea is that lower level modules in the brain continually try to predict what signals they will next receive; and it is only when the predictions fail that signals are passed up to higher (more interconnected) modules, and it is at the highest level they become conscious thoughts. I read his book On Intelligence a few years ago and recommend it. It is quite interesting and has some nice ideas on how to implement an AGI. It conveys a lot of information on neuroscience with a strong focus on the visual cortex. He even derives a machine learning model from his ideas, and it appears to have practical applications: http://en.wikipedia.org/wiki/Hierarchical_temporal_memory On the other hand, the excitement appears to have fizzled out after some initial hype, but maybe I'm just unaware of further progress. Will look into it. Pro: I really appreciate his it's time for computer science to tackle the brain attitude, focused on actually building things; Con: He dismisses the mind-body problem by essentially claiming that consciousness doesn't even exist. It annoys me, but I can tolerate it because he delivers interesting ideas and models on the practical side of things. Not only I am unable to go to that page, but my browser get really mad after I tried. Actually, I have lost my connection with the net. I guess my OS is too old. Strange, it's a regular wikipedia page. Maybe you can try the mobile version, which is lighter: http://en.m.wikipedia.org/wiki/Hierarchical_temporal_memory But from what you say, we might agree on this. That theory is close to the idea that consciousness is called for when automated part of the brain don't fit with the situation. We can open everyday our door without conscious thinking, but if the key appears to not function well, we get conscious of the situation. Right, but it is presented simply as a theory of intelligence. One of the things he mentions in the book, that I really like, is the possibility that the neocortex uses very long axons that go deep into the brain and back in a loop as a device to delay signals. The hypothesis is that it allows the mapping of temporal phenomena into a simpler atemporal pattern-matching problem. Then, if he claims that consciousness does not exist, well, we have a problem. Agreed. Perhaps the problem is not so serious because his theories do not depend on this assertion. It's more of a side remark, he seems to be annoyed by the consciousness issue and just avoids it. It's fair enough, I think. I don't have the book with me so I can't check. I will check in the summer if nobody else comes forward. Telmo. Bruno Best, Telmo. Brent -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. http://iridia.ulb.ac.be/~marchal/ -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
Re: numenta
On 03 Jun 2014, at 16:35, Telmo Menezes wrote: On Tue, Jun 3, 2014 at 4:17 PM, Bruno Marchal marc...@ulb.ac.be wrote: On 02 Jun 2014, at 19:37, Telmo Menezes wrote: On Mon, Jun 2, 2014 at 7:24 PM, meekerdb meeke...@verizon.net wrote: http://numenta.org/ An organization formed by Jeff Hawkins (inventor of the Palm Pilot) to study artificial intelligence. Hawkins idea is that lower level modules in the brain continually try to predict what signals they will next receive; and it is only when the predictions fail that signals are passed up to higher (more interconnected) modules, and it is at the highest level they become conscious thoughts. I read his book On Intelligence a few years ago and recommend it. It is quite interesting and has some nice ideas on how to implement an AGI. It conveys a lot of information on neuroscience with a strong focus on the visual cortex. He even derives a machine learning model from his ideas, and it appears to have practical applications: http://en.wikipedia.org/wiki/Hierarchical_temporal_memory On the other hand, the excitement appears to have fizzled out after some initial hype, but maybe I'm just unaware of further progress. Will look into it. Pro: I really appreciate his it's time for computer science to tackle the brain attitude, focused on actually building things; Con: He dismisses the mind-body problem by essentially claiming that consciousness doesn't even exist. It annoys me, but I can tolerate it because he delivers interesting ideas and models on the practical side of things. Not only I am unable to go to that page, but my browser get really mad after I tried. Actually, I have lost my connection with the net. I guess my OS is too old. Strange, it's a regular wikipedia page. Maybe you can try the mobile version, which is lighter: http://en.m.wikipedia.org/wiki/Hierarchical_temporal_memory May be, tomorrow. Now it works fine. But with mobile version, my non updated old OS makes weird things. But from what you say, we might agree on this. That theory is close to the idea that consciousness is called for when automated part of the brain don't fit with the situation. We can open everyday our door without conscious thinking, but if the key appears to not function well, we get conscious of the situation. Right, but it is presented simply as a theory of intelligence. One of the things he mentions in the book, that I really like, is the possibility that the neocortex uses very long axons that go deep into the brain and back in a loop as a device to delay signals. The hypothesis is that it allows the mapping of temporal phenomena into a simpler atemporal pattern-matching problem. OK. Nice. Then, if he claims that consciousness does not exist, well, we have a problem. Agreed. Perhaps the problem is not so serious because his theories do not depend on this assertion. It's more of a side remark, he seems to be annoyed by the consciousness issue and just avoids it. It's fair enough, I think. Oh yes. No problem as long as it does not mock the question publicly, or eliminate consciousness and person without saying. I don't have the book with me so I can't check. I will check in the summer if nobody else comes forward. OK. I am a bit overwhelmed by the information (like everybody I guess), so I prioritize with respect of what seems the most fun (divine or cosmic fun perhaps) to me. Nice sum up like this are useful, thanks. Bruno Telmo. Bruno Best, Telmo. Brent -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything- l...@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything- l...@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. http://iridia.ulb.ac.be/~marchal/ -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are
numenta
http://numenta.org/ An organization formed by Jeff Hawkins (inventor of the Palm Pilot) to study artificial intelligence. Hawkins idea is that lower level modules in the brain continually try to predict what signals they will next receive; and it is only when the predictions fail that signals are passed up to higher (more interconnected) modules, and it is at the highest level they become conscious thoughts. Brent -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
Re: numenta
On Mon, Jun 2, 2014 at 7:24 PM, meekerdb meeke...@verizon.net wrote: http://numenta.org/ An organization formed by Jeff Hawkins (inventor of the Palm Pilot) to study artificial intelligence. Hawkins idea is that lower level modules in the brain continually try to predict what signals they will next receive; and it is only when the predictions fail that signals are passed up to higher (more interconnected) modules, and it is at the highest level they become conscious thoughts. I read his book On Intelligence a few years ago and recommend it. It is quite interesting and has some nice ideas on how to implement an AGI. It conveys a lot of information on neuroscience with a strong focus on the visual cortex. He even derives a machine learning model from his ideas, and it appears to have practical applications: http://en.wikipedia.org/wiki/Hierarchical_temporal_memory On the other hand, the excitement appears to have fizzled out after some initial hype, but maybe I'm just unaware of further progress. Will look into it. Pro: I really appreciate his it's time for computer science to tackle the brain attitude, focused on actually building things; Con: He dismisses the mind-body problem by essentially claiming that consciousness doesn't even exist. It annoys me, but I can tolerate it because he delivers interesting ideas and models on the practical side of things. Best, Telmo. Brent -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups Everything List group. To unsubscribe from this group and stop receiving emails from it, send an email to everything-list+unsubscr...@googlegroups.com. To post to this group, send email to everything-list@googlegroups.com. Visit this group at http://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.