Transfer Learning using MNIST

One of the most powerful tools in Deep Learning is that sometimes we can take the knowledge or parameters the neural network has learned from one task and apply that knowledge to a different task. So for example maybe we have a neural network model, learned to recognize objects like cats, dogs, and other animals. Then we use that knowledge or use a part of it to do a better job at reading X-ray scans. This is called Transfer Learning. To have a more concrete definition, in transfer learning we reuse a pre-trained model on a new problem.