ai-algorithm.md

ai-algorithm.md

Algorithm Type	Algorithm	Description
Supervised Learning	Linear Regression	Models the relationship between a dependent variable and one or more independent variables.
	Logistic Regression	Used for binary classification problems, predicting probabilities of categorical outcomes.
	Decision Trees	Tree-like model of decisions and their possible consequences, including outcomes and costs.
	Random Forest	Ensemble method using multiple decision trees to improve accuracy and control over-fitting.
	Support Vector Machines (SVM)	Finds the hyperplane that best separates classes in a feature space.
	K-Nearest Neighbors (KNN)	Classifies based on the majority class among the k nearest neighbors in the feature space.
	Naive Bayes	Probabilistic classifier based on Bayes' theorem with strong independence assumptions.
	Neural Networks	Composed of interconnected nodes (neurons), useful for complex pattern recognition tasks.
	Gradient Boosting Machines (GBM)	Ensemble method using boosting to create a strong model from many weak models.
	AdaBoost	Boosting algorithm that combines weak classifiers to form a strong classifier.
Unsupervised Learning	K-Means Clustering	Partitions data into k clusters, where each data point belongs to the cluster with the nearest mean.
	Hierarchical Clustering	Builds a hierarchy of clusters using a bottom-up or top-down approach.
	Principal Component Analysis (PCA)	Reduces the dimensionality of the data while retaining most of the variance.
	Independent Component Analysis (ICA)	Separates a multivariate signal into additive independent non-Gaussian signals.
	DBSCAN	Density-Based Spatial Clustering of Applications with Noise, finds clusters of varying shapes.
	Autoencoders	Neural networks used to learn efficient codings of input data.
	t-Distributed Stochastic Neighbor Embedding (t-SNE)	Reduces dimensionality for visualization of high-dimensional data.
Reinforcement Learning	Q-Learning	Model-free reinforcement learning algorithm to learn a policy for a Markov Decision Process (MDP).
	Deep Q-Networks (DQN)	Combines Q-Learning with deep neural networks for high-dimensional state spaces.
	SARSA	State-Action-Reward-State-Action, updates policy based on the action taken.
	Policy Gradient Methods	Optimizes the policy directly by gradient ascent on expected reward.
Semi-Supervised Learning	Co-Training	Trains classifiers on two different views of the same data, using labeled and unlabeled data.
	Self-Training	Uses its own predictions to label unlabeled data iteratively.
Ensemble Learning	Bagging	Combines predictions from multiple models to reduce variance.
	Stacking	Combines multiple classifiers using a meta-classifier.
Deep Learning	Convolutional Neural Networks (CNN)	Specialized for processing grid-like data, such as images.
	Recurrent Neural Networks (RNN)	Suitable for sequential data, such as time series or natural language.
	Long Short-Term Memory Networks (LSTM)	A type of RNN capable of learning long-term dependencies.
	Generative Adversarial Networks (GAN)	Pits two neural networks against each other to generate realistic data.
	Transformer Networks	Designed for handling sequential data, particularly effective in NLP tasks.
Dimensionality Reduction	Singular Value Decomposition (SVD)	Factorizes a matrix into three matrices, used in PCA and other techniques.
	Linear Discriminant Analysis (LDA)	Finds a linear combination of features that best separates two or more classes.
Optimization	Genetic Algorithms	Search heuristic that mimics the process of natural selection.
	Simulated Annealing	Probabilistic technique for approximating the global optimum of a given function.
Probabilistic Methods	Markov Chains	Models transitions from one state to another in a probabilistic manner.
	Hidden Markov Models (HMM)	Models sequences of observable events generated by hidden states.
	Bayesian Networks	Probabilistic graphical models representing a set of variables and their conditional dependencies.