The validation loss is decreasing very slowly and the accuracy after 100 epochs is significantly lower than the train accuracy. What seems to be the problem here and how can I fix it? Do I decrease the number of nodes in the hidden layer, or use dropout?
Asked
Active
Viewed 9 times
0
-
[What should I do when my neural network doesn't generalize well?](https://stats.stackexchange.com/questions/365778/what-should-i-do-when-my-neural-network-doesnt-generalize-well) – user2974951 May 20 '21 at 11:22
-
You can try reducing the complexity of your model, and use regularization. – gunes May 20 '21 at 11:23