Classification of Machine Learning Algorithms

Preface

Machine learning originated from artificial intelligence, which can give computers the ability that traditional programming can not achieve, such as autopilot of aircraft, face recognition, computer vision and data mining.

There are many machine learning algorithms. What puzzles people most of the time is that many algorithms are one kind of algorithms, and some of them are extended from other algorithms. Here, we will introduce you from two aspects, the first is the way of learning, the second is the similarity of the algorithm.


learning style

Classifying the algorithm according to the learning method can enable people to consider the most appropriate algorithm according to the input data to obtain the best results when modeling and selecting the algorithm.

Supervised Learning

Classification of Machine Learning Algorithms

In supervised learning, input data is called "training data", and each group of training data has a clear class label. When establishing the prediction model, supervised learning establishes a learning process, compares the prediction results with the actual results of "training data", and constantly adjusts the prediction model until the prediction results of the model reach a desired accuracy.

Common application scenarios of supervised learning, such as classification problems and regression problems. Common algorithms are Linear Regression, LogisTIc Regression, Neural Network, SVMs.

Unsupervised Learning

Classification of Machine Learning Algorithms

In unsupervised learning, data is not specifically identified, and the learning model is to infer some internal structure of data.

Common application scenarios include association rules learning and clustering. Common algorithms include K-means Clustering, Principal Component Analysis and Anomaly DetecTIon.

Semi-supervised learning

Classification of Machine Learning Algorithms

In this learning mode, the input data part is marked and the part is not marked. This learning model can be used for prediction, but the model first needs to learn the intrinsic structure of the data in order to organize the data reasonably for prediction. Application scenarios include classification and regression, and algorithms include extensions of commonly used supervised learning algorithms. These algorithms first attempt to model unlabeled data, and then predict the identified data. Such as Graph Inference or Laplacian SVM.

Reinforcement Learning

Classification of Machine Learning Algorithms

In Reinforcement Learning, input data serves as feedback to the model, unlike supervised models, input data serves only as a way of checking model errors. In reinforcement learning, input data is directly fed back to the model, which must be adjusted immediately. Common application scenarios include dynamic systems and robot control. Common algorithms include Q-Learning and Temporal Difference Learning.

In the context of enterprise data application, the most commonly used model may be supervised learning and unsupervised learning. In the field of image recognition, semi-supervised learning is a hot topic because of the existence of a large number of non-identifiable data and a small number of identifiable data. And reinforcement learning is more used in robotic control and other areas that need system control.

Algorithmic Similarity

Regression algorithm

Classification of Machine Learning Algorithms

Regression algorithm is a kind of algorithm that tries to explore the relationship between variables by measuring the error. Regression algorithm is a powerful tool for statistical machine learning. In the field of machine learning, people talk about regression, sometimes referring to a class of problems, sometimes referring to a class of algorithms, which often confuses beginners. Common regression algorithms include: Ordinary Least Square, Logistic TIc Regression, Stepwise Regression, MulTI Variate Adaptive Regression Splines, and Locally Estimated Scatterplot Smoot Estimation. Hing).

Kernel-based algorithm

Classification of Machine Learning Algorithms

The most famous kernel-based algorithm is Support Vector Machine (SVM). Kernel-based algorithms map input data to a higher-order vector space, in which some classification or regression problems can be solved more easily. Common core-based algorithms include: Support Vector Machine (SVM), Radial Basis Function (RBF), and Linear Discriminate Analysis (LDA).

clustering algorithm

Classification of Machine Learning Algorithms

Clustering, like regression, sometimes describes a class of problems, sometimes describes a class of algorithms. Clustering algorithm usually merges the input data according to the central point or hierarchical way. So clustering algorithms are trying to find the intrinsic structure of data in order to classify the data according to the greatest common ground. Common clustering algorithms include k-Means algorithm and Expectation Maximization (EM).

Dimension Reduction Algorithms

Classification of Machine Learning Algorithms

Like clustering algorithms, dimensionality reduction algorithms attempt to analyze the intrinsic structure of data, but dimensionality reduction algorithms attempt to use less information to induce or interpret data in an unsupervised way. Such algorithms can be used to visualize high-dimensional data or to simplify data for supervised learning. Common algorithms include: Principle Component Analysis (PCA), Partial Least Square Regression (PLS).

Integrated Learning

Please read the Chinese version for details.