This is the clearest explanation of SVM optimization math-wise. Other channels just don’t have the patience or the partial derivative skill to clarify the details of every step.
Excellent simplification of a complex concept...surprising that there are so very views...in fact one of the best explanations i have come across...thank you for your efforts...
I found this the best explanation on the internet! Still, I have one question: The primal optimazition is minimizing the weight vector. At the dual optimization are we minimaizing lamdas aswell?
Thank you. To your question: If you are using the dual formulation, you are optimising the lambdas. In the cost function are only lambdas. Then, you can directly calculate the weights with w=sum lambda*x*y
![Kernel Trick for Support Vector Machines [Lecture 3.4]](http://i.ytimg.com/vi/6fqONx4mMI8/mqdefault.jpg)
![Kernel Trick for Support Vector Machines [Lecture 3.4]](/img/tr.png)
I have searched all over the internet and this is by far the best explanation of the derivation of the Dual formula for an SVM.
This is the clearest explanation of SVM optimization math-wise. Other channels just don’t have the patience or the partial derivative skill to clarify the details of every step.
Excellent simplification of a complex concept...surprising that there are so very views...in fact one of the best explanations i have come across...thank you for your efforts...
I found this the best explanation on the internet! Still, I have one question: The primal optimazition is minimizing the weight vector. At the dual optimization are we minimaizing lamdas aswell?
Thank you. To your question: If you are using the dual formulation, you are optimising the lambdas. In the cost function are only lambdas. Then, you can directly calculate the weights with w=sum lambda*x*y
Thank you so much! This really helped me!
hello, what do the subscript i and j represent in the double summation?
The subsricpts i and j represent summations over the N data points to train the model.