Questions

1. What happens if the input values are not scaled in the input dataset?
2. What could happen if the background has a white pixel color while the content has a black pixel color when you're training a neural network?
3. What impact does the batch size have on the model's training time, as well as its accuracy over a given number of epochs?
4. What impact does the input value range have on the weight distribution at the end of the training?
5. How does batch normalization help improve accuracy?
6. How do we know if a model has overfitted on training data?
7. How does regularization help in avoiding overfitting?
8. How do L1 and L2 regularization differ?
9. How does dropout help in reducing overfitting?