New answers tagged convolutional-neural-network
0
votes
Does it make sense to regularize the loss function for binary/multi-class classification?
A lot of times in NN training, we put L2 regularization in the optimizer (called weight decay) instead of on loss function
0
votes
Variational Autoenconder reconstruction loss dominates kl loss
In my scenario, I have over a thousand features, and the kl loss was drowned out by the reconstruction loss. One thing that seems to be working for me is to scale the reconstruction loss based on the ...
- 1
4
votes
why am I getting worst results when using CNN for feature extraction and SVM for classification
There is no one method that always works best. (Google the "No Free Lunch" theorem.)
That means that, conversely, different tools work best in different situations.
And that of course also ...
- 131k
Top 50 recent answers are included