daviddoji/pbdl-book
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

smaller updates to figures and captions

Browse Source
This commit is contained in:
NT
2021-04-01 16:53:41 +08:00
parent 1eba53dca5
commit 17389f35a3
5 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
diffphys-discuss.md
View File

@@ -49,7 +49,7 @@ In this case we compute a (potentially very long) sequence of PDE solver steps i
```{figure} resources/diffphys-multistep.jpg
---
height: 220px
height: 180px
name: diffphys-mulitstep
---
Time stepping with interleaved DP and NN operations for $k$ solver iterations.
@@ -59,7 +59,7 @@ Note that this picture (and the ones before) have assumed an _additive_ influenc
DP setups with many time steps can be difficult to train: the gradients need to backpropagate through the full chain of PDE solver evaluations and NN evaluations. Typically, each of them represents a non-linear and complex function. Hence for larger numbers of steps, the vanishing and exploding gradient problem can make training difficult (see {doc}`diffphys-code-sol` for some practical tipps how to alleviate this).
## Alternatives - Noise
## Alternatives: Noise
It is worth mentioning here that other works have proposed perturbing the inputs and
the iterations at training time with noise {cite}`sanchez2020learning` (somewhat similar to