AI (Artificial Intelligence) common algorithms of artificial intelligence and machine learning
This article is reprinted: The article found in order to understand the AI of MPP.
One, artificial intelligence algorithm
Individual Artificial Intelligence
- Random -
A* Search Algorithm
-Finite-State Machine
- Decision Tree- Game Theroy
- Artificial Neural Networks
- Deep Learning
- Belief Network
- Convolutional Neural Network
- Annealing Algorithm (Simulate Anneal Arithmetic) //empty
group artificial intelligence
- Genetic Algorithm (Genetic Algorithm)
- Cluster technology //empty
Second, the common algorithm of machine learning
Machine learning is undoubtedly a hot topic in the current data analysis field. Many people use machine learning algorithms more or less in their daily work. Here, IT Manager Network summarizes common machine learning algorithms for your reference in work and study. 
There are many algorithms for machine learning. Many times people are confused, many algorithms are a class of algorithms, and some algorithms are extended from other algorithms. Here, we will introduce to you from two aspects, the first aspect is the way of learning, and the second aspect is the similarity of algorithms.
learning method
There are different ways to model a problem depending on the type of data. In the field of machine learning or artificial intelligence, people first think about how algorithms learn. In the field of machine learning, there are several main ways of learning. It is a good idea to categorize algorithms according to how they learn, so that people can consider when modeling and algorithm selection can choose the most appropriate algorithm to obtain the best results based on the input data.
Supervised Learning:
Under supervised learning, the input data is called "training data", and each set of training data has a clear identification or result, such as "spam" and "non-spam" in the anti-spam system, and for handwritten digit recognition. "1", "2", "3", "4" etc. When building a prediction model, supervised learning establishes a learning process, compares the prediction result with the actual result of the "training data", and continuously adjusts the prediction model until the prediction result of the model reaches an expected accuracy rate. Common application scenarios of supervised learning are classification problems and regression problems. Common algorithms include Logistic Regression and Back Propagation Neural Network
Unsupervised Learning:
In unsupervised learning, the data is not specifically identified, and the model is learned to infer some intrinsic structure of the data. Common application scenarios include the learning of association rules and clustering. Common algorithms include Apriori algorithm and k-Means algorithm.
Semi-supervised learning:
In this learning method, the input data is partially marked and partially unmarked. This learning model can be used to make predictions, but the model first needs to learn the internal structure of the data in order to organize the data reasonably to make predictions. Application scenarios include classification and regression, and algorithms include some extensions to commonly used supervised learning algorithms that first attempt to model unlabeled data and then make predictions on labeled data. Graph Inference or Laplacian SVM.
Reinforcement learning:
In this learning mode, the input data is used as feedback to the model. Unlike supervised models, the input data is only used as a way to check whether the model is right or wrong. In reinforcement learning, the input data is directly fed back to the model, and the model must Make adjustments now. Common application scenarios include dynamic systems and robot control. Common algorithms include Q-Learning and Temporal difference learning
In the scenario of enterprise data application, the most commonly used models may be supervised learning and unsupervised learning. In areas such as image recognition, semi-supervised learning is currently a hot topic due to the existence of a large amount of non-identifiable data and a small amount of identifiable data. And reinforcement learning is more used in robot control and other fields that require system control.
Algorithmic Similarity
According to the similarity of the functions and forms of the algorithms, we can classify the algorithms, such as tree-based algorithms, neural network-based algorithms, and so on. Of course, the scope of machine learning is huge, and some algorithms are difficult to categorize clearly into a certain category. For some categories, algorithms for the same category can address different types of problems. Here, we try to classify the commonly used algorithms in the easiest way to understand them.
Regression algorithm:
Regression algorithms are a class of algorithms that attempt to explore relationships between variables using a measure of error. Regression algorithms are a powerful tool in statistical machine learning. In the field of machine learning, when people talk about regression, sometimes they refer to a class of problems, and sometimes they refer to a class of algorithms, which often confuses beginners. Common regression algorithms include: Ordinary Least Square, Logistic Regression, Stepwise Regression, Multivariate Adaptive Regression Splines, and Locally Smoothed Estimation (Locally) Estimated Scatterplot Smoothing)
Instance-Based Algorithms
基于实例的算法常常用来对决策问题建立模型，这样的模型常常先选取一批样本数据，然后根据某些近似性把新数据与样本数据进行比较。通过这种方式来寻找最佳的匹配。因此，基于实例的算法常常也被称为“赢家通吃”学习或者“基于记忆的学习”。常见的算法包括 k-Nearest Neighbor(KNN), 学习矢量量化（Learning Vector Quantization， LVQ），以及自组织映射算法（Self-Organizing Map ， SOM）
正则化方法
正则化方法是其他算法（通常是回归算法）的延伸，根据算法的复杂度对算法进行调整。正则化方法通常对简单模型予以奖励而对复杂算法予以惩罚。常见的算法包括：Ridge Regression， Least Absolute Shrinkage and Selection Operator（LASSO），以及弹性网络（Elastic Net）。
决策树学习
决策树算法根据数据的属性采用树状结构建立决策模型， 决策树模型常常用来解决分类和回归问题。常见的算法包括：分类及回归树（Classification And Regression Tree， CART）， ID3(Iterative Dichotomiser 3)， C4.5， Chi-squared Automatic Interaction Detection(CHAID), Decision Stump, 随机森林（Random Forest）， 多元自适应回归样条（MARS）以及梯度推进机（Gradient Boosting Machine， GBM）
贝叶斯方法
贝叶斯方法算法是基于贝叶斯定理的一类算法，主要用来解决分类和回归问题。常见算法包括：朴素贝叶斯算法，平均单依赖估计（Averaged One-Dependence Estimators， AODE），以及Bayesian Belief Network（BBN）。
基于核的算法
基于核的算法中最著名的莫过于支持向量机（SVM）了。 基于核的算法把输入数据映射到一个高阶的向量空间， 在这些高阶向量空间里， 有些分类或者回归问题能够更容易的解决。 常见的基于核的算法包括：支持向量机（Support Vector Machine， SVM）， 径向基函数（Radial Basis Function ，RBF)， 以及线性判别分析（Linear Discriminate Analysis ，LDA)等
聚类算法
聚类，就像回归一样，有时候人们描述的是一类问题，有时候描述的是一类算法。聚类算法通常按照中心点或者分层的方式对输入数据进行归并。所以的聚类算法都试图找到数据的内在结构，以便按照最大的共同点将数据进行归类。常见的聚类算法包括 k-Means算法以及期望最大化算法（Expectation Maximization， EM）。
关联规则学习
关联规则学习通过寻找最能够解释数据变量之间关系的规则，来找出大量多元数据集中有用的关联规则。常见算法包括 Apriori算法和Eclat算法等。
人工神经网络
人工神经网络算法模拟生物神经网络，是一类模式匹配算法。通常用于解决分类和回归问题。人工神经网络是机器学习的一个庞大的分支，有几百种不同的算法。（其中深度学习就是其中的一类算法，我们会单独讨论），重要的人工神经网络算法包括：感知器神经网络（Perceptron Neural Network）, 反向传递（Back Propagation）， Hopfield网络，自组织映射（Self-Organizing Map, SOM）。学习矢量量化（Learning Vector Quantization， LVQ）
深度学习
Deep learning algorithms are the development of artificial neural networks. It has won a lot of attention recently, especially after Baidu has also begun to develop deep learning, it has attracted a lot of attention in China. In a world where computing power is becoming increasingly cheap, deep learning attempts to build much larger and more complex neural networks. Many deep learning algorithms are semi-supervised learning algorithms designed to process large datasets with small amounts of unlabeled data. Common deep learning algorithms include: Restricted Boltzmann Machine (RBN), Deep Belief Networks (DBN), Convolutional Network, Stacked Auto-encoders.
Dimensionality reduction algorithm
Like clustering algorithms, dimensionality reduction algorithms attempt to analyze the internal structure of the data, but dimensionality reduction algorithms attempt to generalize or interpret data using less information in an unsupervised learning fashion. Such algorithms can be used to visualize high-dimensional data or to simplify data for use in supervised learning. Common algorithms include: Principal Component Analysis (PCA), Partial Least Square Regression (PLS), Sammon Map, Multi-Dimensional Scaling (MDS), Projection Pursuit Wait.
Integration algorithm:
Ensemble algorithms independently train on the same samples with some relatively weak learning models, and then combine the results to make an overall prediction. The main difficulty of the ensemble algorithm lies in which independent and weak learning models to integrate and how to integrate the learning results. This is a very powerful class of algorithms, but also very popular. Common algorithms include: Boosting, Bootstrapped Aggregation (Bagging), AdaBoost, Stacked Generalization (Blending), Gradient Boosting Machine (GBM), Random Forest.