Given a training set DN = xi , yi j i = 1, 2, N with xi 2 Rn and yi

Given a training set DN =



¹xi , yiº j i = 1, 2,


N

with xi 2 Rn and yi 2 f+1, ????1g for all i, assume we want to use a quadratic function y = x|Ax + b|x +

c, where A 2 Rnn, b 2 Rn, and c 2 R, to map from each input xi to each output yi in DN, often called quadratic regression. Derive the closed-form formula to estimate all parameters



A,

b, c

based on the least-square-error criterion.