Answer the "Learning Slowdown" Problem from NNDL Ch. 3: We've discussed at length the learning slowdown that can occur when output neurons saturate, in networks using the quadratic cost to train. Another factor that may inhibit learning is the presence of the x;- term in Equation {61). Because of this term, when an input x;- is near to zero, the corresponding weight wj will learn slowly. Explain why it is not possible to eliminate the xj term through a clever choice of cost function. Problem 0 We've discussed at length the learning slowdown that can occur when output neurons saturate, in networks using the quadratic cost to train. Another factor that may inhibit learning is the presence of the 2:3; term in Equation [61). Because of this term, when an input 223- is near to zero, the corresponding weight wj will learn slowly. Explain why it is not possible to eliminate the aej term through a clever choice of cost function