AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Importance of Computable and Uncomputable Models
The world seems to be computable I mean we keep coming up with mathematical descriptions of relationships between things and they seem consistent or at least locally consistent. The reason it's impossible to work out with the smallest possible representation might be is just because if I'm trying to compute programs that generate regenerate this data I'm trying to compress losslessly so I come up with a program and I run it, see if it regenerates the data that could just run forever. So there are versions of IACSI we can approximate to an arbitrary degree and experiment with. It comes back to the holding problem I can't guarantee I would sort of stop at some point but then I don't know maybe