11 批量/单样本迭代算法 Algorithm 3 (Batch Perceptron) 1 begin initialize a,n(),criterion 0,k =0 2 do kk+1 3 a-a+)∑y y∈Jyk 4 untl(k)∑y<9 y∈Jyk 5 return a 6 end Algorithm 4(Fixed-increment single-sample Perceptron begin initialize a,=0 1 2 dok←(k+1)modn 3 if yk is misclassified by a then a-a-yk 4 until all patterns properly classified 5 return a 6 end11 批量/单样本迭代算法