2 ms·
I know next to nothing about physics, but I do know some about ML. I think the reasons tensor networks are unexplored is interesting: The tools and techniques
by nl 7y ago
I know next to nothing about physics, but I do know some about ML.
I think the reasons tensor networks are unexplored is interesting: The tools and techniques for dealing with them build on those for neural networks and the theoretical benefits are not clear cut enough to gain them a foothold over the practical results of neural networks.
- miles7 7y agoAgreed. There is still a lot of work left to do!
- miles7 7y agoForgot to mention, but here is a recent theoretical result (prediction of generalization performance giving size of training set) based on tensor networks: https://itensor.org/miles/GenerativeMPS.pdf https://itensor.org/miles/GenerativeMPS.pdf