Support Vector Machines (SVM) Hard Margin Dual Formulation - Math Explained Step By Step

This video is a summary of math behind dual formulation of Hard Margin Support Vector Machines (SVM).
Get ready for your interviews understanding the math behind algorithms.
My AI and Generative AI Courses are details here:
ai.generativem...
To get a FREE invite to our LIVE classes, fill below link:
invite.generat...

[@devimam5569]

Nice explanation. Please illustrate, how we can calculate the values of alpha from the dual formula, As we don't have any prior knowledge about which data points are support vectors and which are not.

[@machinelearningmastery]

Its an optimization of iteratively learning the support vectors as we enforce the constraint. This is why gradient descent type optimizer are so slow to converge and typically not used in implementation. You can see this link for more details : www.robots.ox.ac.uk/~az/lectures/ml/lect3.pdf

[@purvanyatyagi2494]

Hi can you tell me what that does I, j corrospond to 2 different points in the data set or something else.......

[@machinelearningmastery]

Yes, that's correct. The additional indexes shows up only in Dual Formulation since we substituted w as a linear equation of alpha,x and y. A clear write up on the same is here: See Page 2 & Page 3. www.robots.ox.ac.uk/~az/lectures/ml/lect3.pdf

[@mayureshsawalwade9800]

can u please share pdf of this?