AI Glossary: Master 50+ Essential AI Terms
Navigate the world of AI with confidence. Explore AIPediaHub's comprehensive glossary featuring over 50 key AI terms, tools, and concepts.
- By Max Maserman
- June 20, 2024
To get a thorough understanding of how many of the following terms work together, be sure to read our AI Fundamentals article.
Action Space
In reinforcement learning, the set of all possible actions that an agent can take in a given environment.
Algorithm
A set of rules or instructions given to an AI system to help it learn from data.
Autoencoder
A type of neural network used to learn efficient codings of unlabeled data, typically for dimensionality reduction.
Autonomous
Machines or systems capable of performing tasks without human intervention.
Backward Chaining
A reasoning method that starts with the goal and works backwards to deduce the facts that support the goal.
Bagging
Short for Bootstrap Aggregating, it’s a machine learning ensemble technique that improves the stability and accuracy of machine learning algorithms by combining multiple models.
Bias
A systematic error in data or the AI model’s predictions, which can lead to unfair or skewed outcomes.
Big Data
Extremely large data sets that may be analyzed computationally to reveal patterns, trends, and associations.
Boosting
A machine learning ensemble meta-algorithm that primarily reduces bias, and also variance in supervised learning, and is used to convert weak learners to strong ones.
Bounding Box
In computer vision, a rectangular box used to define the location of a target object in images.
Chatbot
A software application used to conduct an on-line chat conversation via text or text-to-speech, in lieu of providing direct contact with a live human agent.
Classification
A machine learning task of predicting the class or category of given input data.
Clustering
A type of unsupervised learning method that involves the grouping of data points. In theory, data points that are in the same group should have similar properties and/or features, while data points in different groups should have highly dissimilar properties and/or features.
Cognitive Computing
A subset of AI that attempts to mimic human thought processes in a computerized model.
Cognitive Computing
Similar to artificial intelligence, cognitive computing focuses on mimicking human decision-making processes in a computerized model.
Computer Vision
An interdisciplinary field that deals with how computers can be made to gain high-level understanding from digital images or videos.
Computational Learning Theory
A subfield of machine learning concerned with algorithms’ quantitative properties and capabilities.
Convolutional Neural Network (CNN)
A deep learning algorithm which can take in an input image, assign importance (learnable weights and biases) to various aspects/objects in the image and be able to differentiate one from the other.
Corpus
A large collection of texts, speeches, or other data used to build and train language models.
Cross-validation
A technique for assessing how the results of a statistical analysis will generalize to an independent data set. It is mainly used in settings where the goal is prediction, and one wants to estimate how accurately a predictive model will perform in practice.
Data Mining
The practice of examining large pre-existing databases in order to generate new information.
Data Science
A field that uses scientific methods, processes, algorithms, and systems to extract knowledge and insights from structured and unstructured data.
Dataset
A collection of data specifically prepared to train or test an AI model.
Decision Tree
A decision support tool that uses a tree-like model of decisions and their possible consequences, including chance event outcomes, resource costs, and utility.
Deep Dream
A computer vision program created by Google engineer Alexander Mordvintsev which uses a convolutional neural network to find and enhance patterns in images via algorithmic pareidolia.
Deep Learning
A subset of machine learning involving neural networks with many layers that learn progressively higher-level features in data.
Dimensionality Reduction
The process of reducing the number of random variables under consideration, by obtaining a set of principal variables.
Discount Factor
In reinforcement learning and decision theory, the factor by which future rewards are diminished as compared to immediate rewards.
Dynamic Programming
A method for solving complex problems by breaking them down into simpler subproblems, solving each of those subproblems just once, and storing their solutions.
Entity Annotation
The process of marking up a text to identify and label data structures such as the names of persons, organizations, or locations.
Entity Extraction
The process in NLP of identifying and classifying key elements from text into predefined categories.
Ensemble Learning
The process by which multiple models, such as classifiers or experts, are strategically generated and combined to solve a particular computational intelligence problem.
Episode
A sequence of states, actions, and rewards, which ends with a terminal state in a reinforcement learning scenario.
Evolutionary Computation
A subfield of artificial intelligence (AI) involving combinatorial optimization problems, which are solved with heuristic algorithms based on evolutionary biology such as reproduction, mutation, recombination, and selection.
Expert System
An artificial intelligence program that has expert-level knowledge about a particular domain and knows how to use its knowledge to respond properly.
Exploration-Exploitation Tradeoff
A dilemma faced by algorithms that must choose between exploiting known resources and exploring unknown possibilities to find more valuable resources.
NOTE: Why unstructured data? Unstructured data (such as raw text from websites, books, and articles, or images from the internet) is more abundant and inherently diverse — providing a wealth of human knowledge, language, and visual information. This diversity is crucial for developing models with a broad understanding and the ability to generalize across a wide range of tasks.
Face Recognition
A type of technology capable of identifying or verifying a person from a digital image or a video frame from a video source, typically using features from the face.
Feature Engineering
The process of using domain knowledge of the data to create features that make machine learning algorithms work.
Feature Selection
The process of selecting a subset of relevant features for use in model construction.
Fine-Tuning
A process to refine the weights of a network trained on a specific task to make it perform better in a new task.
Forward Chaining
A method in artificial intelligence used to reason from known facts to an unknown through a series of logical steps.
Function Approximation
A method used in machine learning where a function is approximated from a set of data through which the machine learns.
Fuzzy Logic
A form of many-valued logic in which the truth values of variables may be any real number between 0 and 1. It is used to handle the concept of partial truth, where the truth value may range between completely true and completely false.
GAN (Generative Adversarial Network)
A class of machine learning frameworks designed by Ian Goodfellow and his colleagues, where two neural networks contest with each other in a game (in the sense of game theory, often but not always in the form of a zero-sum game).
General AI
Artificial intelligence systems that possess the ability to understand, learn, and apply knowledge in a wide variety of contexts.
Generative Model
A model for randomly generating observable data values, typically given some hidden parameters.
Genetic Algorithms
A search heuristic that mimics the process of natural selection and is used to generate solutions to optimization and search problems by relying on bio-inspired operators such as mutation, crossover, and selection.
Heuristics
Rules of thumb that help in simplifying and solving complex decision-making problems, often at the cost of accuracy or completeness.
Hyperparameter
Parameters whose values are used to control the learning process and must be set before training a model.
Hyperparameter Tuning
The process of finding the optimal combination of hyperparameters that minimizes a predefined loss function to improve the performance of a model.
Image Annotation
The process of associating an entire image or segments of an image with a label, often used in training computer vision models.
Image Captioning
The process of generating a textual description of an image using artificial intelligence techniques.
Image Classification
The process of categorizing and labeling groups of pixels or vectors within an image based on specific rules.
Image Colorization
The use of algorithms to add color to monochrome images.
Image Enhancement
The process of improving the quality of a digital image by manipulating the image with software.
Image Generation
The process of creating new images, often from a textual or noise-based input, using machine learning models, especially generative models like GANs.
Image Preprocessing
Operations on images at the lowest level of abstraction whose aim is to improve the image data (features) suppressing undesired distortions or enhancing some image features important for further processing.
Image Restoration
The operation of taking a corrupted/noisy image and estimating the clean, original image.
Image Retrieval
The process of searching for and retrieving images from a large database based on features such as color, shape, and texture.
Image Segmentation
The process of partitioning a digital image into multiple segments (sets of pixels, also known as image objects) to simplify or change the representation of an image into something that is more meaningful and easier to analyze.
Image-to-Image Translation
A task where the goal is to learn the mapping between an input image and an output image using a training set of aligned image pairs.
Inference
The process of using a trained model to make predictions.
Inpainting
A technique used to reconstruct lost or deteriorated parts of images and videos.
Intent
In NLP, the goal or purpose behind a user’s input or query.
NOTE: Perplexity is also the name of one of the leading large language models (LLMs).
K-means
A popular clustering algorithm that partitions a set of data points into K clusters, where each data point belongs to the cluster with the nearest mean.
Knowledge Representation
An area of artificial intelligence research aimed at representing information about the world in a form that a computer system can utilize to solve complex tasks.
Label
In machine learning, a tag or category assigned to a piece of data.
Language Model
A statistical model that describes the probability of a token (such as a word or next word in a sequence), based on the previous tokens in a form of text.
Large Language Model (LLM)
A type of language model that is trained on a vast dataset and can handle a broad range of AI tasks, such as text generation, translation, and summarization.
Linguistic Annotation
The process of adding information to a text about its linguistic structure or semantic/syntactic attributes.
Machine Intelligence
Another term for artificial intelligence, emphasizing the machine’s capability to mimic human intelligence.
Machine Learning
A subset of AI that enables systems to learn and improve from experience without being explicitly programmed.
Machine Translation
The application of computers to the task of translating texts from one language to another.
Markov Decision Process
A mathematical framework for modeling decision making in situations where outcomes are partly random and partly under the control of a decision maker.
Markov Property
The memorylessness property of a stochastic process, where the future is independent of the past, given the present.
Markov Reward Process
A Markov Decision Process with rewards and transitions but no actions, used to model sequential decision making where outcomes are partly random.
Model
In AI, a representation (usually mathematical) that a machine learning algorithm uses to make predictions.
Model Deployment
The method by which a machine learning model is integrated into an existing production environment to make practical business decisions based on data.
Model Evaluation
The process of evaluating the performance of a model on a specific set of test data, typically using metrics such as accuracy, precision, recall, F1 score, etc.
Model Selection
The process of selecting one final machine learning model from among a set of candidate models for deployment.
Monte Carlo Method
A broad class of computational algorithms that rely on repeated random sampling to obtain numerical results, typically used to simulate and optimize complex systems.
Multi-armed Bandit
A problem in which a fixed limited set of resources must be allocated between competing (alternative) choices in a way that maximizes their expected gain, based on feedback associated with each of the choices.
Naive Bayes Classifier
A simple probabilistic classifier based on applying Bayes’ theorem with strong (naive) independence assumptions between the features.
Natural Language Generation (NLG)
The use of AI to generate text or speech from a machine representation system such as a knowledge base or a logical form.
Natural Language Processing (NLP)
A field of AI that gives machines the ability to read, understand, and derive meaning from human languages.
Natural Language Understanding (NLU)
The ability of an AI system to understand and interpret human language as it is spoken or written.
Neural Network
A network or circuit of neurons, or in a modern sense, an artificial neural network composed of artificial neurons or nodes.
Normalization
A preprocessing step that changes the values in a dataset to a common scale, without distorting differences in the range of values.
Object Detection
The computer technology related to the identification of specific objects within an image or video, along with their localization through bounding boxes.
Ontology
In the context of AI, an ontology represents structured sets of terms and concepts within a domain and the relationships between them, used to help systems understand and process complex information.
OpenAI
An AI research and deployment company that aims to ensure that artificial general intelligence benefits all of humanity.
Optimal Policy
In decision-making and AI, especially in reinforcement learning, the strategy or set of rules that defines the best decision for every possible state in order to maximize the expected return or minimize risk.
Optimal Value Function
A function that provides the maximum achievable reward obtainable following any policy from each state in a Markov decision process.
Overfitting
A modeling error which occurs when a function is too closely fit to a limited set of data points, affecting the model’s performance on new data.
NOTE: While ‘transformer’ refers to the type of neural network model, the term ‘transformer architecture’ refers to the overall structure and specific components that make the transformer model function (i.e., how data flows through the model, how information is processed and transformed, and how different parts interact to achieve specific tasks).
Parameter
Internal configurations variables of a model that are learned from data.
Pattern Recognition
The recognition of patterns and regularities in data, often used in AI for classification and identification.
Planning
In AI, the process by which an agent programs actions or a sequence of actions to achieve a specific goal.
Policy Iteration
A method used in dynamic programming and reinforcement learning to compute the optimal policy and value function through iterative evaluation and improvement of an initial policy.
Predictive Analytics
The use of data, statistical algorithms, and machine learning techniques to identify the likelihood of future outcomes based on historical data.
Preprocessing
The techniques applied to raw data before feeding it into a machine learning or data processing algorithm, including normalization, scaling, encoding, and handling of missing data.
Prompt
In AI, especially in models like ChatGPT, a user’s input that the model responds to.
Python
A programming language that is widely used in AI due to its readability and robust libraries.
Q-learning
A model-free reinforcement learning algorithm that learns the value of an action in a particular state of a Markov decision process using the Q-function, which essentially gives the expected utility of taking a given action in a given state and following the optimal policy thereafter.
Reasoning
The mental process of deriving logical conclusions and making predictions from available information, extensively simulated in AI through various logical and machine learning techniques.
Recurrent Neural Network (RNN)
A class of artificial neural networks where connections between nodes form a directed graph along a temporal sequence, allowing it to exhibit temporal dynamic behavior.
Regression
A type of predictive modelling technique which investigates the relationship between a dependent (target) and independent variable(s) (predictor).
Reinforcement Learning
An area of machine learning concerned with how intelligent agents ought to take actions in an environment so as to maximize some notion of cumulative reward.
Reinforcement Learning From Human Feedback (RLHF)
A machine learning method where a reinforcement learning model is trained with rewards derived from human feedback to improve its performance.
Reward Function
In AI, particularly in reinforcement learning, a function that maps a state (or a state and action pair) to a reward, which quantifies the value of the outcome.
Robotics
The branch of technology that deals with the design, construction, operation, and application of robots, often incorporating AI methods to enable autonomous function.
SARSA (State-Action-Reward-State-Action)
An algorithm in reinforcement learning that uses the next state and the next action to update its Q-values, unlike Q-learning which uses the maximum reward of the next state.
Semantic Annotation
The process of attaching to a text metadata about its semantic elements for better machine interpretation.
Semantic Web
An extension of the World Wide Web through standards by the World Wide Web Consortium (W3C) which promotes common data formats and exchange protocols on the Web, making web content machine-readable.
Sentiment Analysis
The use of natural language processing to systematically identify, extract, quantify, and study affective states and subjective information.
Stable Diffusion
A type of model that generates high-quality images based on textual descriptions, using a technique known as diffusion models.
Standardization
A preprocessing technique where features are scaled so that they have the properties of a standard normal distribution with a mean of zero and a standard deviation of one.
State Space
The set of all possible states in which a system can exist; in AI, particularly in modeling environments for reinforcement learning.
State Transition
A change from one state to another in a dynamic system, often modeled in stochastic and reinforcement learning environments.
Strong AI
An AI system with generalized human cognitive abilities such that it can solve any problem that a human can solve using intelligence.
Style Transfer
The technique of applying the style of one image to the content of another, used in deep learning fields to merge two images artistically.
Super-resolution
A class of techniques in image processing that enhance the resolution of an imaging system.
Supervised Learning
A type of machine learning algorithm that is trained using labeled data.
Test Data
Data that has never been seen by the model during training, used to evaluate its performance.
Temporal Difference Learning
A class of model-free reinforcement learning methods that learn by bootstrapping from the current estimate of the value function.
Text-to-image model
A model that generates images from textual descriptions, using techniques from machine learning and artificial intelligence.
Trajectory
In AI and robotics, the path that a moving object follows through space as a function of time.
Training Data
Data used to train machine learning models.
Transfer Learning
A research problem in machine learning that focuses on storing knowledge gained while solving one problem and applying it to a different but related problem.
Turing Test
A test of a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human.
NOTE: While ‘transformer’ refers to the type of neural network model, the term ‘transformer architecture’ refers to the overall structure and specific components that make the transformer model function (i.e., how data flows through the model, how information is processed and transformed, and how different parts interact to achieve specific tasks).
Underfitting
A modeling error in machine learning that occurs when a model is too simple to learn the underlying pattern of the data.
Unsupervised Learning
A type of machine learning algorithm used to draw inferences from datasets consisting of input data without labeled responses.
Validation Data
A subset of data used to provide an unbiased evaluation of a model fit on the training dataset while tuning model hyperparameters.
Value Iteration
A method of computing the optimal policy and the value function for a Markov decision process by iteratively improving the value function estimates.
Variance
The variability of model prediction for a given data point or a value indicating the spread of a data distribution.
Variational Autoencoder (VAE)
A type of autoencoder that generates compact representations or encodings of data, used for tasks such as image generation and anomaly detection.
Weak AI
AI systems that are designed and trained for a particular task (virtual personal assistants, such as Apple’s Siri).