Gradient-Sensitive Optimization for Convolutional Neural Networks

Figure 6

Beale’s function. The parameters are iterated 1,000 times. The results are shown in Figure 7 and Table 2.

(a)

(b)