Some related information around the paper... I talked with the main author, Bruno, at the BICA conference last week. We are wondering if we could collaborate.
As for question answering systems with RNN, I found Yann LeCun's introduction in his slides (pp. 18-31): https://drive.google.com/file/d/0BxKBnD5y2M8NbWN6XzM5UXkwNDA/view?pli=1 At the BICA conference, I also met other researchers working on the emergent approach for language such as Peter Ford Dominey, whose works include: "Recurrent temporal networks and language acquisition— from corticostriatal neurophysiology to reservoir computing," (2013) http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3733003/ and Ricardo Gudwin, whose presentation on "Evolving conceptual spaces for symbol grounding in language games" was quite interesting (seems not on the Web yet), comparable to Luc Steels' works. (His team is making an open cognitive architecture framework: http://cst.fee.unicamp.br/ too.) Besides BICA, I found a survey on the emergent approach for language in robotics today: http://arxiv.org/pdf/1509.08973v1.pdf -- Naoya A. > 2015-11-15 13:04, Logan Streondj <[email protected]> wrote: > > Yes, I've always thought of the neuro-net/evolutionary approaches to be the > "meat" of AGI or soft-AI, whereas the structures/classical-AI as the "bones" > of AGI or hard-AI. > > Worms and mollusks are alive and are largely boneless, > though vertebrate sophistication does need bones. > > Anyways my wonder is how to integrate this with OpenCog, or SPEL. > > SPEL is a very hard system at the moment, which makes it finicky like a > compiler, > so I think it would be great to have soft-AI at the edges where interacting > with humans, > That way it can make assumptions and self-corrections for those who wish to > interact in a "softer" more ambiguous fashion. > > Another place for this soft-AI that learns gradually would be inside a > companion-AGI, which I believe will be one of the first kinds of widely-used > AGI's, as they are a natural extension of smart-phones. The companion-AGI > would at first know little about it's user, but gradually learn more, and > adapt to a users quirks. Thus their own phone after awhile of use would be > easier to use than a brand new one of the same type, even if it had all the > same apps and settings. So forwards compatibility would be necessary so that > users don't have to retrain new companion-AI's each time they get a new > smart-phone, which companies love. It will provide for more ways of > extracting money from users, for instance cloud extensions and backups of the > phone AI. > > Some thoughts, > Logan > > > > On Sat, Nov 14, 2015 at 9:57 PM, Ben Goertzel <[email protected]> wrote: > The paper is here... http://arxiv.org/abs/1506.03229 > > Sensationalist media article here: > http://www.iflscience.com/technology/scientists-create-artificial-system-capable-learning-human-language > > This is from Angelo Cangelosi (among others), who works with the iCub robot > and gave a keynote at AGI-12 at Oxford... > > It's very good stuff, but unlike what that news article says, this is not the > first time automated response-generation has been done w/ neural nets.... I > recall a paper by some Russian dude giving similar results in the "Artificial > Brains" special issue of Neurocomputing that Hugo DeGaris and I co-edited > some years ago... > > What distinguishes this work is more the sophistication of the underlying > cognitive architecture ... maybe it works better than prior NNs trained for > dialogue-response or maybe it doesn't; careful comparison isn't given > (understandably -- there is no standard test corpus for this stuff, and prior > researchers mostly didn't open their code).... But the cognitive > architecture is very carefully constructed in a psychologically realistic > way; combined with the interesting practical results, this is pretty nifty... > > The training method is interesting, incrementally feeding the system facts > with increasing complexity, while interacting with it along the way, and > letting it build up its knowledge bit by bit. A couple weeks ago I talked > to a Russian company at RobotWorld in Korea who was training a Russian NLP > dialogue system in a similar way.... (again with those Russians!!) > > Note that with this method, the system can respond to questions involving the > word "dad" without really knowing what a "dad" is (e.g. without knowing that > a dad is a human or is older than a child, etc.). This is just fine, and > people can do this too. But we should avoid assuming that just because it > gives responses that, if heard from a human, would result from a certain sort > of understanding, the system is demonstrating that same sort of > understanding. This system is building up question-response patterns from > the data fed into it, and then performing some generalization. The AI > question is whether the kind of generalization it is performing is really the > right kind to support generally intelligent cognition. > > My thought is that the kind of processing their network is doing, actually > plays only a minor supporting rule in human question-answering and dialogue > behavior. They are using a somewhat realistic cognitive architecture for > reactive processing, and a somewhat realistic neural learning mechanism -- > but the way the learning mechanism is used within the architecture for > processing language, is not very much like the way the brain processes > language. The consequence of this difference is that their system is not > really forming the kinds of abstractions that a human mind (even a child's > mind) automatically forms when processing this kind of linguistic > information.... The result of this is that the kinds of question-answering, > question-asking, concept formation etc. their system can do will not actually > resemble that of a human child, even though their system's answer-generation > process may, under certain restrictions, give results resembling those you > get from a human child... > > These observations do not really contradict anything they say in the paper, > at least upon my quick read.... > > An interesting step, anyway... > > > -- > Ben Goertzel, PhD > http://goertzel.org > > "The reasonable man adapts himself to the world: the unreasonable one > persists in trying to adapt the world to himself. Therefore all progress > depends on the unreasonable man." -- George Bernard Shaw ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657 Powered by Listbox: http://www.listbox.com
