On Tue, Feb 16, 2021 at 7:45 AM Ben Goertzel <[email protected]> wrote:
> One more twist re Bellman/Schrodinger: the representation of dynamic > programming in terms of Galois connections from > > https://www.sciencedirect.com/science/article/pii/S1567832612000525 > > which lets us map dynamic programming into hylomorphisms (or > chronomorphisms if we want to take into account caching for efficiency > optimization), then on the quantum level seems to correspond to the > Consistent Histories interpretation, i.e. the forward and backward > operators composes in the consistent-histories class operator > correspond to folding and unfolding in the hylomorphism representation > of the standard recursive dynamic programming algorithm... > > Fun! > It's very frustrating that it's been almost 100 years (1926) since Schodinger first wrote down that equation, and we have still not cracked its mystery... The path integral is a big step forward, equating the propagator as U(t) = exp(iS/h) with the utility function S. But the path integral is itself a postulate that requires further explanation. If we cannot fully elucidate what is going on in QM, then we have no way to apply QM to dynamic programming even though there is a strong heuristic correspondence between the two. Well, I can think of a "shot in the dark" by simply transitioning the AI problem to the QM version and running the algorithm to see if it outperforms the classical version. It would be fun to try, but it's not exact science 😆 ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T54594b98b5b98f83-Me8546b243f4b4e243ac9d509 Delivery options: https://agi.topicbox.com/groups/agi/subscription
