Introduction to ML

Categories of Machine Learning

Supervised learning

: from labeled data

Unsupervised learning

: from unlabeled data
➡️ Clustering ( 비슷한것 끼리 묶음 ), reproduction ( auto-encoder ), latent variable models ( hidden factor analysis )

Semi-supervised learning

: Small amount of labeled data + Large amount of unlabeled data

Reinforcement learning

: State, Action, Reward
➡️ Learn actions to maximize reward

Recognition System

Overfitting

: 너무 복잡한 Model에 나타남 ➡️ 더 많은 Data로 충분한 학습 필요
: Training error decreases, but Test error increases

Bayesian Theorem