Lecture 13 - AsyDynamics/CS231n Wiki

Unsupervised learning

Supervised learning

Data(x,y); x is data, y is label Goal: train function, to map from x to y

Unsupervised learning

Data, no label; cheap! Goal: learn underlying hidden structure of the data Example: Clustering, dimensionality reduction, feature learning, density estimation

Generative model

Given training data, generate new samples from same distribution

Variational Autoencoders VAE