Types of Machine Learning
Download as PPTX, PDF•6 likes•15,794 views
The document discusses four main types of machine learning: supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning. Supervised learning uses labeled data for classification and prediction, while unsupervised learning processes unlabeled data for tasks like clustering and anomaly detection. Semi-supervised learning combines both labeled and unlabeled data, and reinforcement learning involves agents learning through interactions with their environment to maximize rewards.
Types of Machine Learning
- 2. TYPES OF MACHINE LEARNING Machine Learning algorithm are divided into four types, based on the following criteria. Whether or not they are trained with human supervision 1. Supervised Learning 2. Unsupervised Learning 3. Semi supervised Learning 4. Reinforcement Learning
- 3. SUPERVISED LEARNING In supervised learning, the training set that provides data for an algorithm contains a convenient solution called labels. A typical supervised learning task is classification. Spam filter is a good example. It has been trained on with sample emails along with a label defined (spam or ham). After learning from existing emails it must classify new emails. Another common task is prediction for example car price prediction. For prediction it offers a number of features like mileage, age, make, etc. called predictors. With the help of predictors new car prize will be predicted. That kind of work called regression.
- 4. SUPERVISED LEARNING ALGORITHMS k-Nearest Neighbors Linear Regression Logistic Regression Support Vector Machines (SVMs) Decision Trees and Random Forests Neural networks
- 5. UNSUPERVISED LEARNING In Unsupervised Learning training data is unlabeled. In this type of machine learning, system learn without a teacher. Some important Unsupervised Learning algorithm are given below: 1. Clustering(K-Means) 2. Anomaly detection and novelty detection (One-class SVM, Isolation Forest) 3. Visualization and dimensionality reduction (PCA) 4. Association rule learning
- 6. SEMI- SUPERVISED LEARNING Labelling data is very costly and time-consuming process. Some algorithms can deal with data that’s partially labeled means few labeled instances and few unlabeled instances. This is called semi-supervised learning. Some photo-hosting services, such as Google Photos, are good examples of this. Semi-supervised learning algorithms are combinations of unsupervised and supervised algorithms e.g. deep belief networks (DBNs).
- 7. REINFORCEMENT LEARNING Reinforcement learning is a whole other animal. This learning system is known as Agent. In reinforcement learning an action is taken by observing the environment and in return get reward. It learns by itself which is best policy to get the most reward over time. For example, many robots implement Reinforcement Learning algorithms to learn how to walk. DeepMind’s AlphaGo program is also a good example of Reinforcement Learning
- 8. SOURCE Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow Author: Aurélien Géron
- 9. Thank You