Many people use the MSE decomposition to illustrate bias and variance. However, is there any statistical learning theory connecting these concepts? Namely, is there a formula calculating model capacity using bias and variance?
$\begingroup$
$\endgroup$
3
-
1$\begingroup$ Qualitatively, models that are richer / more complex have higher variance and lower bias. $\endgroup$– angryavianOct 27, 2020 at 21:55
-
$\begingroup$ what is model capacity $\endgroup$– develaristOct 27, 2020 at 23:50
-
$\begingroup$ @angryavian Is there any equation depicting such kind of relationship? $\endgroup$– JP ZhangOct 29, 2020 at 0:47
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
I think this issue is perfectly discussed and explained by Dr. Kilian Weinberger from Cornel University.
Here are the lecture notes: http://www.cs.cornell.edu/courses/cs4780/2018fa/lectures/lecturenote12.html
and http://www.cs.cornell.edu/courses/cs4780/2018fa/lectures/lecturenote11.html
You can also find video recordings of the course, so you can follow up.
