[telchar]

And yet neural networks can solve symbolic integral and derivative problems and differential equations better than other computer algebra programs. Sure one network might fail to compute sin(x) numerically, but another could easily tell you its derivative is cos(x). Turns out they are pretty flexible. Do they need to do everything?

[ABeeSea]

Source? I would be very surprised if there was a neural symbolic PDE solver better than what’s in wolfram mathematica.

[SemanticStrengh]

Related https://arxiv.org/abs/1806.07366

[mhh__]

At the cost of how many parameters?