New answers tagged loss-functions
2
votes
Accepted
Custom Loss function Overfits to the Wrong Output but MSE Doesn't
The problem is that the loss function only has 0 gradient when the neural network output is 0. This is not desirable! Instead, we want a loss function that has a 0 gradient when the output matches the ...
Top 50 recent answers are included
Related Tags
loss-functions × 1174neural-networks × 371
machine-learning × 351
regression × 138
classification × 120
optimization × 116
cross-entropy × 77
gradient-descent × 60
logistic × 55
regularization × 47
tensorflow × 47
convolutional-neural-network × 47
boosting × 45
keras × 45
bayesian × 42
decision-theory × 42
autoencoders × 38
python × 33
svm × 32
self-study × 30
maximum-likelihood × 29
accuracy × 28
predictive-models × 27
estimation × 26
mse × 26