3 ms·
I don't think it's a given that the architectures that get us to AGI will be constrained in exactly the way current neural networks are.
by wutbrodo 6y ago
I don't think it's a given that the architectures that get us to AGI will be constrained in exactly the way current neural networks are.
- CodeGlitch 6y agoYes I could be like the difference between birds and planes, still using the laws of aerodynamics but in a different way. One of those scales, the other doesn't.