Steve, >> Certainly one can do probabilistic reasoning about rates of change of >> probabilities. But can you give an example to illustrate why this is a >> profoundly better approach? > > > Because temporal learning, instead of being a challenge requiring some > sort of "workaround", becomes the normal mode of operation that would > require a workaround to avoid doing.
For instance, suppose we want to reason about (or estimate the truth value of) P( Ben and Jade are eating breakfast | Cassio recently ate breakfast) In OpenCog we'd do this in terms of fuzzy temporal predicates describing the events involved (i.e. predicates whose truth is restricted to a certain fuzzy time interval). How is this better done using derivatives of probabilities? Or, for a neural modeling example, consider Beck's paper "Probabilistic Population Codes for Bayesian Decision Making" http://www.google.com/url?sa=t&rct=j&q=pouget%20probabilistic%20population%20coding&source=web&cd=1&cad=rja&ved=0CCcQFjAA&url=http%3A%2F%2Fwww.bcs.rochester.edu%2Fpeople%2Falex%2Fpub%2Farticles%2FBecketalNeuron08.pdf&ei=4MRAUISxA9Tm6gHM94CICQ&usg=AFQjCNGnoF0em0GFizMoPwuoGG2nyvKigw&sig2=5JykWk80s6eRRv2Pdv3F7Q which is related to perceptual decision making in monkeys, and models behavior of neurons in the LIP. How would this model (or others like it) be improved via hypothesizing neurons whose inputs and outputs are the derivatives of probabilities? I'm trying to get a concrete sense of the practical meaning of your suggestion... > That was my thought. They have money, lots of people, and really need the > industry that such a thing would likely create. Meanwhile, research in > America is stone cold dead, and is likely to stay that way for a looooong > time, perhaps forever. I sense a long plane flight in my future. America still leads the world in scientific research, by a long shot. The list of amazing results coming out of US university and company labs never stops, thankfully ;-) ... But America's lead in the sciences is gradually diminishing, and in some fields American science is way more conservative than it should be (due to the locked-in opinions/paradigms of the "research elites" commanding the funding and tenure decisions in these fields).... So there is certainly the possibility of some of the next great breakthroughs coming from elsewhere, Asia being the prime contender.... Knowing they are coming from behind, the Asian nations may be more willing to gamble on speculative/wild ideas -- which is sometimes a prescription for big success... -- Ben >I wonder > what language I should be learning to speak? Putonghua would be my suggestion, if you need to pick just one... But in Hong Kong I'm doing fine with just English ;) ben ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968 Powered by Listbox: http://www.listbox.com
