Slashdot Mirror

← Back to Stories (view on slashdot.org)

Jeff Hawkins' Cortex Sim Platform Available

Posted by kdawson on from the build-a-brain-at-home dept.

UnreasonableMan writes "Jeff Hawkins is best known for founding Palm Computing and Handspring, but for the last eighteen months he's been working on his third company, Numenta. In his 2005 book, On Intelligence, Hawkins laid out a theoretical framework describing how the neocortex processes sensory inputs and provides outputs back to the body. Numenta's goal is to build a software model of the human brain capable of face recognition, object identification, driving, and other tasks currently best undertaken by humans. For an overview see Hawkins' 2005 presentation at UC Berkeley. It includes a demonstration of an early version of the software that can recognize handwritten letters and distinguish between stick figure dogs and cats. White papers are available at Numenta's website. Numenta wisely decided to build a community of developers rather than trying to make everything proprietary. Yesterday they released the first version of their free development platform and the source code for their algorithms to anyone who wants to download it."

11 of 126 comments (clear)

  1. Right... by Bugpowda · · Score: 5, Insightful

    I'm still a bit confused as to how he is so confident that this is how the neocortex works given that this is still one of the 23 unsolved problems in system neuroscience. But hey, he made a lot of money off Palm, that gives him way more street cred than people who have been working on this problem for their whole lives.

    1. Re:Right... by fyngyrz · · Score: 3, Insightful
      his claims were quite, uhm, let's say, "ambitious".

      That is a wonderful thing, though. First of all, claims can be tested. They'll either live up to the description, or they won't. If the don't, another path not to go down in a particular manner has been identified, and that is useful. OTOH, if they are verified, then we may have a key to a form of cognition. Whether it is our kind or not is really not as important as just the fact that it is some kind.

      Aside from that, I found some very interesting things in his descriptions of the HTM. For instance, I found the following precise description of enabling religious behavior: First, he describes how HTMs handle specific, non-overlapping domains (and of course this doesn't mean that another HTM can't relate those to each other.) One might handle financial markets, another speech, another cars. Then he says "After initial training, an HTM can continue to learn or not" Emphasis mine. So you can set up an HTM in a learning situation where you limit the input to descriptions consisting of sensory data of any arbitrarily limited set of patterns you like, get it to see the world represented by those patterns as you wish, and then disable learning for that particular HTM. Other HTMs can continue to learn, but that one is "frozen." Sounds like the perfect recipe for a priest or supplicant to me. Does that not sound like the very core definition of "unshakable faith"?

      For all the doubt being thrown this fellow's way, you know, eventually someone will come up with something like this and it will be a working model of such a system. It's a tough problem, very abstract and requiring a lot of insight, but as with all problems discovered to date where we can actually get our hands on the system under study, there is no indication that any part of it exists in any way outside the sphere of nature and the natural rules we already know - and we know a lot of basic rules.

      Kudos to him for sinking his teeth into the problem, and for coming up with results that can be tested, and for letting them loose into the word for such testing. If he's wrong, he's helping. If he's right - he's going to be mentioned in the same breath with a lot of very important people for a very, very long time to come.

      --
      I've fallen off your lawn, and I can't get up.
    2. Re:Right... by Walt+Dismal · · Score: 5, Interesting

      I've been working for some time on technology with hierarchical NN architecture like Hawkin's HTM, but mine in part involves SIPO FIFOs with attached neural networks, and the output of the NNs go to the next layer's SIPO+NNs, and so on up the chain. It's intended to extract meaning from symbol flow over time. Like speech primitives into language. Hawkins embeds temporal symbol handling in each HTM layer in a different way. Both of us are trying to emulate some of the processing the neocortex does, but I am less concerned with matching closely the brain and more concerned with outperforming the limitations of the brain. I believe there are classes of problems his architecture will solve, but can't handle others. There's lots of room for people to explore what his technology can do, and I expect it will work well for some things.

    3. Re:Right... by Christianson · · Score: 3, Interesting
      Caveat: I am a neuroscientist. I am not familiar with the works of Mr. Hawkins.

      That is a wonderful thing, though. First of all, claims can be tested. They'll either live up to the description, or they won't.

      Most "grand-scale theories of brain operation", in fact, fail to make claims that can be tested, at least not in the foreseeable future. They predict the large-scale algorithms by which the brain operates. They do not make any claims as to the behavior of any individual neurons, and this is the data we have to work with. Moreover, these theories generally fail to provide any explanation for existing data, such as the diversity in neuronal phenotypes, the connectivity architecture, functional segregation, the wealth of neurotransmitters, laminar structure and why the details of this structure varies across the neocortex, differences in histochemical labelling, and so on and so forth. In short, these theories tend to be computer science, and not neuroscience. They might represent major progress in the question, "how do we make a machine that can solve a difficult computational problem?" but they have very little significance in answering the question, "what are the principles that underlie neural performance?"

      Aside from that, I found some very interesting things in his descriptions of the HTM. For instance, I found the following precise description of enabling religious behavior: First, he describes how HTMs handle specific, non-overlapping domains (and of course this doesn't mean that another HTM can't relate those to each other.) One might handle financial markets, another speech, another cars. Then he says "After initial training, an HTM can continue to learn or not" Emphasis mine. So you can set up an HTM in a learning situation where you limit the input to descriptions consisting of sensory data of any arbitrarily limited set of patterns you like, get it to see the world represented by those patterns as you wish, and then disable learning for that particular HTM. Other HTMs can continue to learn, but that one is "frozen." Sounds like the perfect recipe for a priest or supplicant to me. Does that not sound like the very core definition of "unshakable faith"?

      Not really. The ability to stop learning is a crucial element of learning. In the existing computational literature, this is related to the problem of overfitting: there comes a point where additional learning is dominated by attempts to explain noise in the data, and can actually lead to degraded performance. A classic example of "frozen learning" in living animals include the zebra finch male (who learns one song only in his life, which remains unchanged past adolescence). Anecedotally, you can also think of human accents in speech; most people never lose the accents they develop in childhood, no matter how hard they might try. Of course, both of the examples illustrate that it is not a matter of stopping learning; learning can in fact occur, but much slower and under much more extreme conditions.

      From the perspective of neuroscience (and in fact, from the unsupervised learning perspective in general), given that there are lots of models that can learn and stop learning, the much more relevant question is: how can the system switch between these two states?

      For all the doubt being thrown this fellow's way, you know, eventually someone will come up with something like this and it will be a working model of such a system. It's a tough problem, very abstract and requiring a lot of insight, but as with all problems discovered to date where we can actually get our hands on the system under study, there is no indication that any part of it exists in any way outside the sphere of nature and the natural rules we already know - and we know a lot of basic rules.

      Is the problem a supernatural one? Of course not. It is a very tough problem. The issue is not, at this point, a lack of theory.

  2. High-Quality Video Link by overeduc8ed · · Score: 5, Informative

    High quality versions of Jeff Hawkin's talk at UC Berkeley are available here.

  3. Re:Barrier to entry by RyanFenton · · Score: 5, Insightful

    Don't be so afraid of complexity - Slashdotters make fun of themselves for diving into things uneducated (not reading the articles, not RTFM), but really, the only way to cope with such an informationaly complex landscape such as computing is to sometimes just be willing to go unprepared and be willing to make mistakes, and to ask stupid questions.

    Not so much dare to be stupid, but rather the Socratic, don't be afraid of exposing your own ignorance - don't lose your opportunity to learn by merely being embarrassed of people thinking you dumb while you take your first few steps in a new landscape.

    But do take notes and research the small topics you are uncertain of after your first adventure into to the topic. Perhaps you'll need to learn a bit about XML/XSL, perhaps you'll need to find out the anatomy of a nerve cell to understand some explanations. If nothing else though - get into it because it is a fun adventure and a lot of cool stuff to learn.

    Ryan Fenton

  4. Re:Barrier to entry by Wagoo · · Score: 4, Informative

    Hawkins' published a book before this was implemented in code called "On Intelligence". You could do worse than starting by reading through that.

    He's also done some lectures available on Google Video.

  5. Confidentiality agreement a killer by else58 · · Score: 5, Informative
    The download license looked fine until the Confidentiality paragraph. Does it really say that anything I learn from Numenta is confidential property of Numenta?

    Confidentiality. 1. Protection of Confidential Information. You agree that all code, inventions, algorithms, business concepts, workflow, ideas, and all other business, technical and financial information, including but not limited to the HTM Algorithms, HTM Algorithms Source Code, and HTM Technology, that you obtain or learn from Numenta in connection with this Agreement are the confidential property of Numenta (Confidential Information). Except as authorized herein, you will hold in confidence and not use, except as permitted or required in the Agreement, or disclose any Confidential Information and you will similarly bind your employees in writing. You will not be obligated under this Section 6 with respect to information that you can document: (i) is or has become readily publicly available without restriction through no fault of you or your employees or agents; or (ii) is received without restriction from a third party lawfully in possession of such information and lawfully empowered to disclose such information; or (iii) was rightfully in your possession without restriction prior to its disclosure by Numenta; or (iv) was independently developed by your employees or consultants without access to such Confidential Information.
  6. Re: Not one year, seven or eight years by stephanruby · · Score: 3, Informative

    That book was published over a year ago, lots can and has changed in that time.

    Actually, its content was produced seven or eight years ago.

    Its publishing date was "December 2005". But publishers will lie about the publication date of a book if it allows them to sell more books. And in this case, I wouldn't be surprised if the book came out hot off the presses in December 2004 with a postdate of "December 2005"

    Furthermore, this book was based on the scientific proceedings of a conference which occurred six years before the book was finally edited (or finally published). I'm actually not sure of the year of the scientific conference itself, because the information supplied to sell the book doesn't give the actual year.

  7. almost... by penguinbroker · · Score: 3, Interesting
    This would be great if computational power was dirt cheap. People smarter then you or i have already thought about this.

    http://en.wikipedia.org/wiki/Baum-Welch_algorithm http://en.wikipedia.org/wiki/Viterbi_algorithm

    The first is an alogorithm which utilizes forward and back-tracking "to find the unknown parameters of a hidden Markov model." The second is a similar algorithm used for learning 'known' causes (for reference).

    I work in computational linguistics and the time an algorithm takes to run and the amount of memory it requires are serious limitations. That's why ad-hoc systems are so common.

  8. Cortex Sim == Bullsh*t by Anonymous Coward · · Score: 5, Interesting

    Hawkins is a rich guy, and no-one feels like telling him that his stuff is crap. He had a few smart people working for him at some point, but when they told him his ideas were half baked and not new, he just fired their asses.

    Here is what many people in machine learning and computer vision think about Hawkins stuff:
    - it's way, way behind what other people in vision and machine learning are doing. Several teams have biologically-inspired vision systems that can ACTUALLY LEARN TO RECOGNIZE 3D OBJECTS. Hawkins merely has a small hack that can recognize stick figures on 8x8 pixel binary images. Neural net people were doing much more impressive stuff 15 years ago.
    - Hawkins's ideas on how the brain learns are not new at all. Many scientists in machine learning, computer vision, and computational neuroscience have had general ideas similar to the ones described in Hawkins's book for a very long time. But scientists never talk about philosophical ideas without actual scientific evidence to support them. So instead of writing popular book with half-baked conceptual ideas, they actually build theories and algorithms, they build models, and they apply them to real data to see how they work. Then they write a scientific paper about the results, but they rarely talk about the philosophy behind the results.

    It's not unusual for someone to come up with an idea they think is brand new and will revolutionize the world. Then they try to turn those conceptual ideas into real science and practical technologies, and quickly realize that it's very hard (the things they thought of as mere details often turn out to be huge conceptual obstacles). Then, they realize that many people had the same ideas before, but encountered the same problems when trying to reduce them to practice (which is why you didn't hear about their/your ideas before). These people eventually scaled back their ambitions and started working on ideas that were considerably less revolutionary, but considerably more likely to result in research grants, scientific publications, VC funding, or revenues.

    Most people go through that "naive" phase (thinking they will revolutionize science) while they are grad students. A few of them become successful scientists. A tiny number of them actually manage to revolutionize science or create new trends. Hawkins quit grad school and never had a chance to go through that phase. Now that he is rich and famous, the only way he will understand the limits of his idea is by wasting lots of money (since he obviously doesn't care about such things as "peer review"). In fact, many reputable AI scientists have made wild claims about the future success of their latest new idea (Newell/Simon with the "general theorem prover", Rosenblatt with the "Perceptron", Papert who thought in the 50's that vision would be solved over the summer, Minsky with is "Society of Minds", etc......).

    No scientist will tell Hawkins all this, because it would serve no purpose (other than pissing him off). And there is a tiny (but non-zero) probability that his stuff will actually advance the field.

        - Anonymous Scientist