Prove Theorem 2, showing that when the activation function is nonlinear, say, yk = (vk), the generalized

Prove Theorem 2, showing that when the activation function is nonlinear, say, yk = φ(vk), the generalized delta rule can be formulated as wkj (n + 1) = wkj (n) − ηgkj , where gkj = (yk − tk)

∂φ

∂vk xj .