Luca Ambrogioni counters that random inputs add entropy enabling novelty.
(((ل()(ل() 'yoav))))👾#95
@yoavgo
These newfangled AI math proofs aren't impressive: all the results are contained within the definitions and axioms!
No Digg Deeper questions have been answered for this story yet.
@yoavgo There’s no new information anyways in anything we do, because of the data processing inequality 😂
where are the information theory people who will tell us there is no new information in the proof because of the data processing inequality?
@LucaAmb regardless of the process it combined known ideas in a known way. you could theoretically run it exhaustively without randomness, it would just take much longer
@yoavgo That does not work because there are random numbers coming it. There is genuinely new entropy involved.
At most you can say there is no information concerning the original training dataset, but that's besides the point
@DimitrisPapail exactly!
@yoavgo There’s no new information anyways in anything we do, because of the data processing inequality 😂
@andrewgwils @EmilevanKrieken @yoavgo Ha I was reading your paper just yesterday!
@EmilevanKrieken @yoavgo It was actually that the DPI doesn’t hold under computational constraints and shouldn’t be used in arguments about what information a learner could extract.
@andrewgwils @EmilevanKrieken yes
@EmilevanKrieken @yoavgo It was actually that the DPI doesn’t hold under computational constraints and shouldn’t be used in arguments about what information a learner could extract.