This forum is inactive. Browsing/searching possible.
Connect your moderator Slack workspace to receive post notifications:
Duality: when is dual better than primal? (unclear)
Lectures
·
62
·
1
·
11 Jan '21
Hi!
I'm still confused as to which cases are better with dual or with primal. The formulation o…
Classifying with the kernel K
Lectures
·
105
·
5
·
29 Dec '20
Hello! I can't understand the passages associated to the topic "Classifying with the kernel K" in l…
Question regarding kernelized ridge regression
Lectures
·
67
·
1
·
28 Dec '20
Good morning, I was a bit confused by how the expression for $$\alpha^{*}$$ was obtained in the alt…
alphas in SVMs
Lectures
·
90
·
2
·
27 Dec '20
Hi,
I was reading over the conditions for alpha in SVM. I understand if it is correctly classifi…
Hard vs Soft Margin SVM interpretation
Lectures
·
76
·
2
·
26 Dec '20
Hello, I have a doubt about the interpretation I should give to hard/soft margin SVM formulations. …
Primal formula
Lectures
·
113
·
7
·
4 Nov '20
Hi,
I don't understand why in the videos we've got this formula
![Screenshot from 2020-11-03 11…
Don't understand this statement: 'Note that G(w, α) is convex in w and linear, hence concave, in α'
Lectures
·
94
·
5
·
31 Oct '20
I am not sure how this conclusion is drawn in the Lecture 7 SVM:
"Note that G(w, α) is convex in w…
soft vs hard margin dual and alpha
Lectures
·
96
·
2
·
30 Oct '20
Good Evening,
I have a hard time seeing how we derive the constraints for alpha in the dual form…
Set of classifiers is not convex
Lectures
·
133
·
6
·
29 Oct '20
Hello,
In the first part of the first lecture (w7), when talking about the convexity of the loss…
Question on Coordinate descent
Lectures
·
93
·
2
·
28 Oct '20
Hi, Why is it easy to optimize via coordinate ascent to solve dual optimization?
Could you explai…
SVM soft margin
Lectures
·
75
·
2
·
28 Oct '20
I was wondering if the penalization for being on the wrong side of the margin (slack variables cost…
SVM Dual problem optimisation
Lectures
·
151
·
4
·
28 Oct '20
When trying to find
$$max_{\alpha\in R^N} : \alpha^T1 - \frac{1}{2\lambda}\alpha YXX^TY\alpha$$…