1.2 Basic Neural Network example: Housing price prediction

• x: input (size, #bedrooms, zipcode, wealth)
• y: result (house price)
• neural network: stacked neurons

x -> neural network -> y

Given enough data for training x and y, neural network are remarkly good at figuring out functions that accurately map from x to y

1.3 Data Source

• Structured data: Database
• Unstructured data: Audio, Image, Text

1.4 Scale drives deep learning progress

• Scale of the data and size of the neural network
• It takes a lot of time to train a net only in large data area, neural network performs better than other approaches

Data, computation, Algorithms

• Change algorithm make computation faster

• idea -> code -> experiment -> idea