a) In Deep Learning Networks what are the aspects you should take into consideration in choosing an activation function ? (i) (ii) b) In light of your answer to (a) which one(s) of the following would you chose as an activation function ? Explain your answer. (i) f(x) = -min(2, x) (ii) f(x) = 0.9x + 1 (iii) f(x) = (iv) f(x) = (v) f(x)= = [min(x,.1x) |x >= 0 min(x, .lx) x < 0 [max(r,.lx) |x>=0 min(x,.1x) x < 0 1 |x >=0 |x