I agree with this Matthew. It's important to remember how much of our behaviour is hard wired into our brains due to evolution. Simpler animals like ants have pretty much everything wired in at birth. Despite their tiny brains (and no neocortex btw) ants are capable of tasks like raft building and farming. If ants have so much pre-wiring you can bet humans have much more. Until we understand how ants can possibly do these tasks (in a very detailed neuron-by-neuron way) we'll never by able to make an artificial insect. If we can't make an insect, we can't make a mouse. If we can't make a mouse we have NO HOPE at an artifical human!
Trying to make a human brain in 2015 is like the Wright Brothers in 1903 trying to make an A380 jumbo jet. They had to start small and simple and slowly work up. Every plane they got to fly gave them skills to develop the next model. The other thing about them is, when they actually got a machine to fly, every layman could tell they had succeeded! For Nupic it has to be the same. We have to eventually produce something that any layman can appreciate is intelligent. In my opinion it must therefore be embodied (virtual reality is OK too). So the question: what is the ultimate goal of Nupic? How will we know we succeeded? We can't just present an animation of active/inactive neurons on a computer screen. Nor is a good fit to a time series enough. Even if we assure people the algorithm is based on the human brain they will not be impressed. That would be like the Wright brothers saying "well we don't have a flying machine -- how vulgar! -- but look at these wonderful equations we wrote! The problem of flight is solved!". No it's got to be something real, which humans can tell is intelligent -- which SCREAMS intelligent! My advice: an artifical ant which can pass the "ant Turing test", i.e. with all the intelligence, responsiveness, intuition, survival instinct etc of a real ant (possibly even accepted by the ant colony). Don't underestimate how much effort this will take! On Tue, Jun 30, 2015 at 9:21 PM, Matthew Lohbihler <[email protected]> wrote: > The Helen Keller example is probably not appropriate since her intelligence > was the result of an evolutionary process that expected inputs typical of an > embodiment, and in the absence of a number of those developed itself with > what input was available. I would say this only indicates how similar humans > really are, not how an intelligence can develop from any old basis. Trying > to create an intelligence that is not embodied the way John suggests might > work in some fashion, but don't expect it to be able to understand or relate > much to humanity. > > > On 6/30/2015 3:40 PM, Julian Samaroo wrote: > > The body is really just the input and output systems that the neural > structures attach to, as well as providing the ever-important task of > supplying nutrients, fighting disease, etc. But in AI, it is simply > abstracted away, and you are left with a set of sensory inputs and motor > outputs, which can be anything you'd like. As David said, sight and audition > aren't vital to intelligence in its most simplistic form, and you can thus > attach something like HTM to any input-output pair that you can think up > (although this doesn't necessarily imply anything useful will come of it). > > Julian Samaroo > Manager of Information Technology > BluePrint Pathways, LLC > (516) 993-1150 > > On Tue, Jun 30, 2015 at 2:36 PM, cogmission (David Ray) > <[email protected]> wrote: >> >> John, >> >> I just thought of this. I wonder how relevant the experiences of sensory >> deprived individuals are to a comparison of the capabilities of >> "dis-embodied" intelligences? Someone like Helen Keller who maybe only had >> kinesthetic and taste senses, could maybe be analogous in some way to a >> developing dis-embodied intelligence? Maybe not, just a thought... >> >> On Tue, Jun 30, 2015 at 2:21 PM, Matthew Taylor <[email protected]> wrote: >>> >>> John, >>> >>> Just to make sure that all your questions have been addressed directly: >>> >>> On Tue, Jun 30, 2015 at 2:55 AM, John Blackburn >>> <[email protected]> wrote: >>> > "performs with true intelligence" is a pretty bold claim. If this is >>> > the case, how come there are no very convincing examples of HTM >>> > working with human like intelligence? The Hotgym example is nice but >>> > it is really no better than what could be achieved with many existing >>> > neural networks. Echo state networks have been around for years and >>> > can make temporal predictions quite well. >>> >>> People define "intelligence" in different ways. If you take for >>> granted that the neocortex has "true intelligence", then HTM might be >>> called an implementation of "true intelligence" algorithms based upon >>> the fact that it acts upon incoming data with the same basic >>> principles as the neocortex. We are trying to lift the intelligence >>> out of the brain and into software, one step at a time. >>> >>> So, while NuPIC's performance might not seem all that impressive when >>> other technologies can do similar things, we have lots of room to grow >>> [1] and a lot more work to do. All of this upcoming work should >>> increase the capabilities of the HTM system we are implementing. The >>> fact that we are somewhat on-par with some other ML techniques at this >>> point is encouraging to me. >>> >>> > I recently presented some >>> > time sequence data relating to a bridge to this forum but HTM did not >>> > succeed in modelling this (ESNs worked much better). >>> >>> I had a little time to work on your bridge tilt data [2], but not >>> enough to make it useful. I still think this problem presents a >>> relevant challenge for HTM, and I think with more time and effort, >>> someone might be able to create a real solution. I, unfortunately, >>> have other projects I have to work on. :( >>> >>> > So outside of >>> > Hotgym, what really compelling demos do you have? I've been away for a >>> > while so maybe I missed something... >>> >>> My current favorites are location-based anomaly demos like these: >>> - https://github.com/nupic-community/mine-hack >>> - https://github.com/numenta/nupic.geospatial >>> >>> I am also working on a new tutorial, coming within a couple weeks >>> (hopefully). >>> >>> > I am also rather concerned HTM needs swarming before it can model >>> > anything. Isn't that "cheating" in a way? It seems the HTM is rather >>> > fragile and needs a lot of help. The human brain does not have this >>> > luxury it just has to cope with whatever data it gets. >>> >>> Swarming is hard to explain. In the brain, input data to the neocortex >>> comes from sensory organs, which have been tuned by millions of years >>> of evolution to have very specific characteristics that process >>> incoming light, sound, movement, etc. into certain patterns of nerve >>> excitations. These patterns get generated outside the cortex, but they >>> are still important to attempt to replicate in some ways. All data in >>> "reality" must be represented to the cortex somehow outside of that >>> reality. In NuPIC, this is what encoders so. They translate data >>> coming into them into a representation similar to a vector of nerve >>> excitations. >>> >>> Anyway, swarming is a very rough way to simulate evolution in the >>> sensory organs. It randomly sets up encoders with different parameters >>> (also spatial pooling and temporal memory parameters) and tries to >>> find the best possible set of configurations for the specific data >>> that is being processed. Your cochlea have had millions of years to >>> come to that perfect set of configuration parameters ;). Swarming is a >>> brute-force attempt to resolve some set of parameters for a specific >>> input data set. It is not always right, it takes a long time, and it >>> sometimes requires manual intervention, but it definitely very useful >>> for finding groups of configurations that work well for certain types >>> of data. >>> >>> > I'm also not convinced the neocortex is everything as Jeff Hawkins >>> > thinks. I seriously doubt the bulk of the brain is just scaffolding. >>> > I've been told birds have no neocortex but are capable of very >>> > intelligent behaviour including constructing tools. Meanwhile I don't >>> > see any AI robot capable of even ant-like intelligence. (ants are >>> > amazing!) Has anyone even constructed a robot based on HTM? >>> >>> While I know nothing about bird brains, except that they have a >>> cerebral cortex that has some similarities to the mammalian cortex, I >>> do know that hierarchy in the neocortex is a generally accepted theory >>> in neuroscience. >>> >>> We could still learn a helluva lot from the lower levels of the brain >>> (imagine a flight vehicle that could control itself as efficiently as >>> a fly), that just isn't what we're trying to do at Numenta. >>> >>> > Personally I don't think a a disembodied computer can ever be >>> > intelligent (not even ant-like intelligence). IMO a robot (and it must >>> > BE a robot) needs to be embodied with sensory-motor loop at the core >>> > of its functionality to start behaving like an animal. >>> >>> You don't need to have physical interaction with the world to have >>> behavior. There are millions of actions that can be taken on the >>> internet that all have consequences, change the landscape for the >>> actor, and present different possible actions in return. The most >>> obvious example is video games, but the internet in general is a very >>> large universe with no physical structure, but endless virtual >>> structures to interact with. >>> >>> [1] https://github.com/numenta/nupic.research/wiki/Current-Research-Tasks >>> [2] https://github.com/nupic-community/bridge-tilt >>> >>> Regards, >>> >> >> >> >> -- >> With kind regards, >> >> David Ray >> Java Solutions Architect >> >> Cortical.io >> Sponsor of: HTM.java >> >> [email protected] >> http://cortical.io > > >
