On Saturday, March 26, 2022, at 3:53 PM, James Bowery wrote: > So, to take the simple example of an f/2 circuit, I'm interested in what > algorithms can, given this (Input, Output) sequence:
This is an interesting question though I couldn’t help thinking about an AGI’s perspective of general circuit optimization. Physical electrical circuits are metagraphs and the resultant virtual electrical behavior seems to be a dynamical system. And there appears to be a many-to-one mapping from the metagraphs to the DS. So a circuit optimizer would create/discover physical circuits of improved graph complexity with the same output DS. And, this optimization can be generalized to other types of networks. Seems though CA’s might be utilizable in the hyperedge/hypernode discovery in digital logic though I’m not sure how it’s typically done there… and/or for more generalized mapping it may be practical to look at various mathematical structure represented in graphs, for example “Groups as Graphs” so perhaps seek correlate mathematical structure that relates to these types of circuits. BTW that online circuit designer/testing app is neat... it appears javascript based. Lately I’ve been trying to think of things by minimizing the arrow-of-time bias. Arrow-of-time seems wrong.. though practical. And the replacement is => ( very interesting ) ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/Tb9b009413931e22d-M6287011ecd6d0b7ebaa2d3e2 Delivery options: https://agi.topicbox.com/groups/agi/subscription
