Well, still similar to the Regularization applied to linear, the regularized cost function for logistic regression is defined by:
$$J(\mathbf{w}, b) = -\frac{1}{m} \sum_{i=1}^m \left[ y^{(i)} \log(f_{\mathbf{w}, b}(\mathbf{x}^{(i)})) + (1 - y^{(i)}) ...
