I was watching this numerical problem on SVM link here. At 4.42 he wrote this equation:

$$ \begin{align} \alpha_1 \tilde{s_1} \cdot \tilde{s_1} + \alpha_2 \tilde{s_2} \cdot \tilde{s_1} + \alpha_3 \tilde{s_3} \cdot \tilde{s_1} &= -1 \\ \alpha_1 \tilde{s_1} \cdot \tilde{s_2} + \alpha_2 \tilde{s_2} \cdot \tilde{s_2} + \alpha_3 \tilde{s_3} \cdot \tilde{s_2} &= +1 \\ \alpha_1 \tilde{s_1} \cdot \tilde{s_3} + \alpha_2 \tilde{s_2} \cdot \tilde{s_3} + \alpha_3 \tilde{s_3} \cdot \tilde{s_3} &= +1 \end{align} $$

where $\tilde{s_1}, \tilde{s_2}, \tilde{s_3}$ are support vectors. But from the support vector we see that: $$ w = \sum \alpha_i y_i s_i$$ where $y_i$ is class label and $s_i$ is support vector. Since $\alpha$'s are zero for all other points except support vector so I am considering only the support vector here. My question is if we look at the above equation he just wrote: $$ w = \sum \alpha_i s_i$$

I can't understand how he wrote $ w = \sum \alpha_i s_i$ and get the correct answer yet?


1 Answer 1


he just wrote: w=∑αisi

I dont think so. He wrote $ \sum_i \alpha_ix_i^Tx$ where $x$ is the test sample. He took $x$ as each support vector and hence that summation will result to 1 or -1 (depending on class).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.