Autoencoder parameter selection in 'trainAutoencoder' function

I was trying to use matlab inbuilt function 'trainAutoencoder'. In autoencoder how i select the mini-batch size(b) and the learning rate. What is the effect of 'L2WeightRegularization', 'SparsityRegularization','SparsityProportion' parameter on the performance.

Answer 1

[Mandar]

Refer to the MATLAB documentation link that discuss the training of the sparse autoencoder networks and the significance of the different training parameters such as 'L2WeightRegularization', 'SparsityRegularization',and 'SparsityProportion' parameter on the performance.

Link: https://in.mathworks.com/help/deeplearning/ref/trainautoencoder.html#buyr01q-1