Machine learning and Deep learning
Introduction
Machine learning
Looking for function
Different types of Functions
Structured Learning
create something with structure(image, document) in other word, learn how to create
How to find a function
1.Function with unknown Parameters
2.Define Loss from Training Data
Loss is a function of paras
Loss: how good a set of values is
If y and y-hat are both probability distributions -> Cross-entropy
3.Optimization
Gradient Descent
local minima
example
Example-> Linear Models
Linear models are too simple … we need more sophisticated modes.
1. Piecewise function
Linear models have severe limitation-> model bias
All piecewise linear curves = constant + sum of set of linear functions
Beyond piecewise linear?
2. The bule lines - sigmoid
3. Different sigmoid
4. Representation
6. Vector representation
7. optimization
8. model optimize
9. fancy name
neuron and neural network
why don’t we go deeper? -> overfitting