It makes perfect sense for different AIs to learn similar representations of the same reality. This is similar to how science works.
@piotr780
3 ай бұрын
this paper kinda proves that reality exist objectively :) btw. there exists interface hypothesis of human perception (that we dont observe objective reality but altered by evolution representation which increase our fitness to environment), if so then mayby our machine are going to observe reality as it is - objectively, but then what about us ? if their perception will be right by "design" and our is not, then it is going to create some kind of tension between us and them - who will be handling who ?
@johnstifter
3 ай бұрын
Thank you for making these, your "Neural Networks as Quantum Field Theories" vid opened the doors to a lot of very interesting avenues of thought. Namely the connection between gravity, holography & quantum entanglement, I think I found something regarding circuit complexity, Geodesic shortest path in the Bloch sphere and something in ML NNs dealing with manifold hypothesis, & information transformations.
@NikolajKuntner
3 ай бұрын
wew
@holthuizenoemoet591
3 ай бұрын
whats up with this thumbnail my dude
@NikolajKuntner
3 ай бұрын
It's the Platonic representation of ML research
@shoopinc
3 ай бұрын
In the Schopenhauerian formulation we might get a nice representation with AI, but we cannot access the will through it.
Пікірлер: 8