Question: In this problem, we use the squared loss ( y,y ) := ( y y ) 2 to evaluate the performance of a neural network.
In this problem, we use the squared loss
(y,y ) := (y y)2
to evaluate the performance of a neural network. Consider the following perceptron parameterized byw:
f(x1,x2) = (w1x1+ w2x2),
where(x) := (1 + ex)1represents the logistic function. Compute the partial gradient ofwith respect tow1at (w1,w2) = (0,0)for the given training sample(x1,x2,y). (Hint: the derivative of(x)is(x) (1 (x)).)
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
