Sequential minimal optimization: Difference between revisions

Content deleted Content added
m Algorithm: - my bad...fixed previous edit
fixed overcounting
Line 52:
# Repeat steps 1 and 2 until convergence.
 
When all the Lagrange multipliers satisfy the KKT conditions (within a user-defined tolerance), the problem has been solved. Although this algorithm is guaranteed to converge, heuristics are used to choose the pair of multipliers so as to accelerate the rate of convergence. This is critical for large data sets since there are <math>n(n-1)/2</math> possible choices for <math>\alpha_i</math> and <math>\alpha_j</math>.
 
== Related Work ==