I was going through the chapter 5 on RPROP in your book "Introduction to Math of Neural Networks" and needed some clarification in the way weight change value was determined as on Page 51.
From what I understood from the theory, if the value of 'c' is zero, the corresponding weight change would be 0. However, on page 51, even though 'c' is zero, weight change is determined as -0.01. It is not clear how this value has been computed because it does not seem to comply with the theory at all.
Is this some initial default value to boost the weight change?