###
Machine Learning Study Note - Introduction

Last updated: 2017-10-10 15:39:14 PDT.

## Types of Machine Learning Problems (wiki)

ML tasks can be classified into three broad categories based on the nature of learning signal or feedback avialable to a learning system:

- Supervised learning
- Unsupervised learning
- Reinforcement learning

Between supervised and unsupervised learning is semi-supervised learning.

Depending on the output:

- Classification
- Regression
- Clustering
- Ranking
- Recommendation

Also:

- Density estimation
- Dimensionality reduction

## Statistics

- Paramatric methods - Methods that make assumptions that the data comes from a distribution and it intends to estimate these parameters.
- Nonparamatric methods - Methods that does not assume certain form of distributions. (Or rather, a free-form of distributions. "nonparametric really means many parametric.").

## Machine Learning Approaches

- Linear Model
- Kernel machines
- Decision tree
- Bayesian Estimation
- Hidden Markov Models
- Ensemble learning
- Random Forest
- Gradient Boosting

- Graphical models (Bayesian networks)
- Clustering
- Association rules learning
- Artificial neural network
- Reinforcement learning
- Generalized Additive Model
- Inductive logic programming
- Feature learning
- Sparse dictionary learning

- Genetic algorithms
- Rule-based machine learning
- Learning classifier systems (LCS)

## Model Selection

Hypotheses set . Bias - underfitting. Variance - overfitting. Generalization. Triple trade-of.

Cross validation. Test set. Analogy with exams.