[SOLVED] ECE472, Deep Learning – Assignment 2

          Problem Statement  Consider a set of examples with two classes and distributions as

 in Figure 1. Given the vector x ∈ R² infer its target class t ∈ {0,1}. As a model 011 use a multi-layer perceptron f which returns an estimate for the conditional ⁰¹² density p(t = 1 | x):

                                                                                                       2

                                                                                                f : R → [0,1]                                                   (1)

       parametrisized by some set of values θ. All of the examples in the training set

should be classified correctedly (i.e. p(t = 1    x) > 0.5 if and only if t = 1).

Impose an L² penalty on the set of parameters. Produce one plot. Show the

 examples and the boundary corresponding to p(t = 1 | x) = 0.5. The plot must be

 of suitable visual quality. It may be difficult to to find an appropriate functional               

 form for f, write a few sentences discussing your various attempts.