Machine Learning Algorithms
There are four main types of learning algorithms:
- Supervised learning algorithm: This is trained to predict an outcome for a given set of input features. It's well studied and widely used in many areas such as spam classification, fraud detection, and product recommendation.
- Unsupervised learning algorithm: This analyzes the underlying patterns or structure of data and groups data into clusters. Examples are outlier detection, fraud detection, and dimensionality reduction.
- Semi-supervised learning: This falls between supervised learning and unsupervised learning. It's intended to boost learning accuracy for a supervised learning model by mixing unlabeled data.
- Reinforcement learning algorithm: This is trained to play a "game." It learns to take a "smarter" action at each step in a game so that it will eventually win the game. Examples are AlphaGo, robot control, and quantitative trading.