New answers tagged overfitting
5
votes
Accepted
Is there an argument against using the (reviewed) predictions of a model as ground truth to further train exactly this model?
Using the (unchecked) predictions of the model as training data is an approach known as "pseudo-labeling". It can help in certain situations, depending on the underlying structure of your ...
1
vote
Accepted
When exactly am I overfitting -- contradicting metrics
From the loss graph I would conclude, that at approx 2k steps overfitting starts, so using the model at approx 2k steps would be the best choice. But looking at the precision graph, training e.g. ...
4
votes
Is there an argument against using the (reviewed) predictions of a model as ground truth to further train exactly this model?
The answer is: It depends.
What you describe is a strategy often used to save time and costs for labelling data. It is important that the data you have already labelled (the 20%) is representative of ...
Top 50 recent answers are included
Related Tags
overfitting × 87neural-networks × 34
machine-learning × 34
deep-learning × 24
convolutional-neural-networks × 12
training × 9
regularization × 8
cross-validation × 8
generalization × 8
reinforcement-learning × 6
accuracy × 6
autoencoders × 5
underfitting × 5
computational-learning-theory × 4
loss × 4
dropout × 4
terminology × 3
keras × 3
datasets × 3
generative-adversarial-networks × 3
deep-neural-networks × 3
hyperparameter-optimization × 3
performance × 3
test-datasets × 3
early-stopping × 3