[jkabrg]

There was a Quanta Magazine article talking about predicting the evolution of a flame-front using ML. The ML algorithm remained accurate for eight Lyapunov intervals; eight times longer than the previous SOTA.

https://www.quantamagazine.org/machine-learnings-amazing-abi...

[foxes]

The only real argument I think you can make is that it might be more efficient to have a neural network quickly spit out an approximate solution instead of solving the actual equations.

But if you have the time having the actual equations is more valuable?

[PeterisP]

Assuming that "having the actual equations" is possible, which in market predictions it most likely is not.