Sparsity regularization

Sparsity regularization

sparsity(high_probability, weight)

Arguments

high_probability

Expected probability of the high value of the encoding layer. Set this to a value near zero in order to minimize activations in that layer.

weight

The weight of the sparsity regularization

Value

A Ruta regularizer object for the sparsity, to be inserted in the encoding layer.

References

• Sparse deep belief net model for visual area V2

• Andrew Ng, Sparse Autoencoder. CS294A Lecture Notes

See also

autoencoder_sparse, make_sparse, is_sparse