Stanford University public Class machine learning: Machines Learning System Design | Data for machine learning (the learning algorithm behaves better when the volume is large)

For the performance of four different algorithms in different size data, it can be seen that with the increase of data volume, the performance of the algorithm tends to be close. That is, no matter how bad the algorithm, the amount of data is very large, the algorithm can perform well.

When the amount of data is large, the learning algorithm behaves better:

Using a larger set of training (which means that it is impossible to fit), the variance will be low, and if many parameters and layers are added to the logistic regression or linear regression model, the deviation will be very low. Together, this will be a good high-performance learning algorithm.

Stanford University public Class machine learning: Machines Learning System Design | Data for machine learning (the learning algorithm behaves better when the volume is large)