Sparse Autoencoder Loss

Problem

Compute the sparse autoencoder objective: mean squared reconstruction error plus l1_coeff times the mean absolute feature activation.

The signature is sparse_autoencoder_loss(x, x_hat, features, l1_coeff).

Hint

The L1 term keeps the dictionary features sparse and interpretable.

case 1

in: [[1,0],[1,0],[0.5,0],0.1]

expected: 0.025

case 2

in: [[1,1],[0,0],[0.5,0.5],1]

expected: 1.5

case 3

in: [[0.5],[0.5],[-0.2],2]

expected: 0.4

case 4

in: [[1,2,3],[1,2,3],[0,0,0],0.5]

expected: 0

Starter codePython

def sparse_autoencoder_loss(x, x_hat, features, l1_coeff):
    # Your code here
    pass

4 test cases · run your code against them in the editor