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

first version DP

Browse Source
This commit is contained in:
NT
2021-01-16 13:30:26 +08:00
parent cbf20ff8fe
commit b68bbcfaa5
5 changed files with 120 additions and 63 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
physicalloss-discuss.md
View File

@@ -16,11 +16,10 @@ And while the setup is realtively simple, it is generally difficult to control.
has flexibility to refine the solution by itself, but at the same time, tricks are necessary
when it doesn't pick the right regions of the solution.
## Is it "Machine Learning"
## Is it "Machine Learning"?
TODO, discuss - more akin to classical optimization:
we test for space/time positions at training time, and are interested in the
solution there afterwards.
we test for space/time positions at training time, and are interested in the solution there afterwards.
hence, no real generalization, or test data with different distribution.
more similar to inverse problem that solves single state e.g. via BFGS or Newton.