A Faster Gradient Ascent Learning Algorithm for Nonlinear SVM

<table>The dependency on <svg height="12.175" id="M187" style="vertical-align:-3.56265pt" version="1.1" viewbox="0 0 8.6374998 12.175" width="8.6374998" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
<g transform="matrix(.017,-0,0,-.017,.062,7.675)"><path d="M478 372q0 -39 -31 -97t-61 -98t-78 -98q-45 -55 -73 -102q-13 -79 -13 -197q-11 -11 -43.5 -25.5t-53.5 -15.5l-15 17q5 35 26 101.5t47 123.5q8 72 -1.5 174.5t-37.5 178.5q-14 37 -29 37q-20 0 -67 -65l-25 21q37 60 73 90.5t63 30.5q47 0 72 -112q13 -56 17.5 -141.5
t0.5 -143.5h2q155 193 155 297q0 26 -12 47q-5 8 -5 15q0 16 12.5 27t29.5 11q21 0 34 -21t13 -55z" id="x1D6FE"></path></g>
</svg> of the number of iterations required by Platt’s SMO and the variants of the gradient ascent algorithms.</table>

International Scholarly Research Notices

fig9

Figure 9

Figure 9: A Faster Gradient Ascent Learning Algorithm for Nonlinear SVM 