Bias vs Variance

underfitting vs overfitting

	Benchmark	High variance	High bias	High variance + high bias
Train set error	0.5%	1%	15%	15%
Dev set (validation) error	1%	11%	16%	30%

If your Neural Network model seems to have high bias, what of the following would be promising things to try?

1. Make the Neural Network deeper

2. Increase the number of units in each hidden layer

-> don't get more training data