100 Artificial Intelligence (AI) Terms You Need to Know

The broad field of computer science dedicated to creating machines that can perform tasks typically requiring human intelligence, such as learning, problem-solving, decision-making, and understanding language.

A subset of AI that enables systems to learn from data without being explicitly programmed. It involves algorithms that can identify patterns, make predictions, and improve their performance over time.

A specialized subfield of Machine Learning that uses artificial neural networks with multiple layers (hence "deep") to learn complex patterns from large amounts of data, often inspired by the structure and function of the human brain.

A computational model inspired by the human brain's structure. It consists of interconnected "neurons" (nodes) organized in layers that process information and learn from data.

A branch of AI that focuses on enabling computers to understand, interpret, and generate human language. This includes tasks like translation, sentiment analysis, and chatbots.

An AI field that allows computers to "see" and interpret visual information from the world, such as images and videos. Applications include facial recognition, object detection, and autonomous vehicles.

A set of well-defined instructions or rules that a computer follows to solve a problem or perform a task. In AI, algorithms are the core logic behind how models learn and make decisions.

A collection of related data used to train, test, and validate an AI or Machine Learning model. The quality and quantity of the data set significantly impact the model's performance.

The portion of the data set used to "teach" an AI model. The model learns patterns and relationships from this data.

A type of machine learning where the model is trained on labeled data, meaning each input in the training set has a corresponding correct output. The model learns to map inputs to outputs.

A type of machine learning where the model learns from unlabeled data, identifying patterns, structures, or relationships within the data without explicit guidance.

A type of machine learning where an agent learns to make decisions by performing actions in an environment and receiving rewards or penalties based on its actions. It learns through trial and error.

In AI, a model is the output of the training process. It's the learned representation of the data that can be used to make predictions or decisions on new, unseen data.

The process of using a trained AI model to make predictions or decisions on new, unseen data. This is when the model is put into "production."

A type of AI that can create new content, such as text, images, audio, or video, that is similar to human-created content but is not a direct copy. Large Language Models (LLMs) are a prime example.

A type of AI model trained on vast amounts of text data to understand, generate, and respond to human language in a coherent and contextually relevant way.

The art and science of crafting effective inputs (prompts) for AI models, especially LLMs, to guide them towards generating desired outputs.

An AI-powered computer program designed to simulate human conversation through text or voice, often used for customer service or information retrieval.

In AI, bias refers to systematic errors or unfairness in a model's predictions or decisions, often stemming from biased training data or algorithmic design.

An approach to AI development that aims to make AI models more transparent and understandable, allowing humans to comprehend how and why a model makes certain decisions.

A hypothetical type of AI that possesses human-level cognitive abilities across a wide range of tasks, rather than being limited to a specific domain.

A hypothetical type of AI that surpasses human intelligence in virtually every field, including scientific creativity, general wisdom, and social skills.

AI designed and trained for a specific task or limited domain, such as playing chess or recommending products. Most current AI systems are narrow AI.

The process of discovering patterns, insights, and knowledge from large datasets, often using machine learning, statistics, and database systems.

An interdisciplinary field that uses scientific methods, processes, algorithms, and systems to extract knowledge and insights from structured and unstructured data.

The process of selecting, transforming, and creating new variables (features) from raw data to improve the performance of machine learning models.

A modeling error where a machine learning model learns the training data too well, including its noise and outliers, leading to poor performance on new, unseen data.

A modeling error where a machine learning model is too simple to capture the underlying patterns in the training data, resulting in poor performance on both training and new data.

Parameters that are set before the training process begins and control the learning process itself (e.g., learning rate, number of layers in a neural network).

An optimization algorithm used to minimize the cost function of a model by iteratively adjusting the model's parameters in the direction of the steepest descent of the cost function.

A type of deep neural network particularly effective for processing grid-like data, such as images, by using specialized layers called convolutional layers.

A type of neural network designed to process sequential data (like text or time series) by maintaining an internal memory of previous inputs.

A neural network architecture that uses self-attention mechanisms to weigh the importance of different parts of the input data, highly effective in NLP tasks and the basis for LLMs.

A component within neural networks that allows the model to focus on specific parts of the input sequence when making predictions, improving performance on long sequences.

Numerical representations of words, phrases, or other data types in a continuous vector space, where similar items are located closer together.

The process of breaking down a text into smaller units called tokens (words, subwords, or characters), which are then processed by NLP models.

The use of NLP to determine the emotional tone or opinion expressed in a piece of text (e.g., positive, negative, neutral).

The ability of a machine or program to identify words spoken aloud and convert them into readable text.

The ability of an AI system to identify and classify objects, people, writing, and actions in images or videos.

A computer vision technique that identifies and locates objects within an image or video, often by drawing bounding boxes around them.

A technology capable of identifying or verifying a person from a digital image or a video frame by comparing selected facial features from the image with faces within a database.

The branch of engineering and computer science that deals with the design, construction, operation, and application of robots, often integrating AI for autonomous behavior.

The use of technology to perform tasks with minimal human intervention, often enhanced by AI to make decisions and adapt to changing conditions.

An early form of AI that mimics the decision-making ability of a human expert in a specific domain, typically using a knowledge base and inference engine.

A problem-solving approach that employs a practical method not guaranteed to be optimal or perfect, but sufficient for reaching an immediate, short-term goal or approximation.

A subfield of AI that aims to simulate human thought processes in a computer model, including reasoning, learning, and interacting with humans naturally.

The use of statistical algorithms and machine learning techniques to identify the likelihood of future outcomes based on historical data.

An AI system that suggests items (products, movies, music, etc.) to users based on their past behavior, preferences, or similarities to other users.

The process of identifying unusual patterns or data points that do not conform to expected behavior, often used in fraud detection or system monitoring.

An unsupervised learning technique that groups data points into clusters based on their similarities, without prior knowledge of the groups.

A supervised learning task where a model learns to categorize new data into predefined classes or categories based on patterns from labeled training data.

A supervised learning task where a model predicts a continuous numerical output (e.g., house prices, temperature) based on input features.

The process of choosing a subset of relevant features (variables) from the original dataset to use in model training, reducing complexity and improving performance.

Techniques used to reduce the number of features or variables in a dataset while preserving its essential information, often to simplify models or visualize high-dimensional data.

The steps taken to clean, transform, and prepare raw data for use in machine learning models, including handling missing values, scaling, and encoding categorical data.

The process of assessing the performance and accuracy of a trained machine learning model using various metrics and techniques (e.g., accuracy, precision, recall, F1-score).

A technique for evaluating machine learning models by training them on different subsets of the data and testing on the remaining unseen data, to ensure robustness and prevent overfitting.

A fundamental concept in machine learning that describes the relationship between a model's ability to fit the training data (bias) and its sensitivity to small fluctuations in the training data (variance).

A function used in neural networks that determines the output of a neuron, introducing non-linearity and allowing the network to learn complex patterns.

An algorithm used to train neural networks by calculating the gradient of the loss function with respect to the weights and biases, and then updating them to minimize the loss.

A function that quantifies the error or difference between the predicted output of a model and the actual target output. The goal of training is to minimize this function.

The process of adjusting a model's parameters to minimize its loss function and improve its performance on a given task.

A hyperparameter in optimization algorithms that determines the step size at each iteration while moving toward a minimum of a loss function.

One complete pass through the entire training dataset during the training of a machine learning model.

The number of training examples utilized in one iteration during the training of a machine learning model.

Techniques used to prevent overfitting in machine learning models by adding a penalty to the loss function for complex models.

A regularization technique used in neural networks where randomly selected neurons are ignored during training, preventing them from co-adapting too much.

A machine learning method where a model trained on one task is re-purposed or fine-tuned for a second related task, leveraging learned features.

The process of taking a pre-trained model and further training it on a smaller, specific dataset to adapt it to a new, related task.

A type of generative AI consisting of two neural networks, a generator and a discriminator, that compete against each other to create realistic synthetic data.

A technique used to align AI models (especially LLMs) with human preferences and values by using human feedback to train a reward model, which then guides the AI's learning.

When an AI model (especially an LLM) generates false, misleading, or nonsensical information that is presented as factual.

Mechanisms or rules implemented to ensure AI systems operate within ethical, legal, and technical boundaries, preventing harmful or undesirable outputs.

A system that can operate independently without continuous human oversight, making decisions and taking actions based on its environment and goals (e.g., self-driving cars).

The deployment of AI and machine learning models directly on edge devices (e.g., smartphones, IoT devices) rather than in the cloud, enabling faster processing and reduced latency.

AI services and infrastructure provided through cloud computing platforms, offering scalable resources for training and deploying AI models.

A set of rules and protocols that allows different software applications to communicate and interact with each other, often used to integrate AI functionalities into other systems.

The process of tagging or marking raw data (images, text, audio) with meaningful labels, making it suitable for supervised machine learning training.

Artificially generated data that mimics the statistical properties of real-world data but does not contain any actual sensitive information, often used for training AI models.

The process of integrating a trained machine learning model into a production environment where it can be used to make predictions or decisions on real-world data.

A set of practices for deploying and maintaining machine learning models in production reliably and efficiently, combining ML, DevOps, and data engineering.

The application of AI and machine learning to automate and enhance IT operations, including monitoring, anomaly detection, and problem resolution.

The use of software robots (bots) to automate repetitive, rule-based digital tasks, often mimicking human interactions with computer systems.

The use of computer systems to assist in the creation, modification, analysis, or optimization of a design, increasingly integrating AI for generative design and optimization.

A simulated experience that can be similar to or completely different from the real world, often enhanced by AI for realistic interactions and adaptive environments.

An interactive experience of a real-world environment where the objects that reside in the real world are "augmented" by computer-generated perceptual information, sometimes powered by AI for object recognition and tracking.

A network of physical objects embedded with sensors, software, and other technologies for the purpose of connecting and exchanging data with other devices and systems over the internet, often generating data for AI analysis.

A virtual representation of a physical object or system, updated in real-time with data from its physical counterpart, allowing for AI-powered simulations and predictive maintenance.

The intersection of quantum computing and artificial intelligence, aiming to leverage quantum phenomena to develop more powerful AI algorithms and models.

A machine learning approach that trains an algorithm across multiple decentralized edge devices or servers holding local data samples, without exchanging data samples.

The degree to which a human can understand the cause of a decision or output of an AI system.

The ability to explain or present in understandable terms how an AI model works internally and makes predictions.

The principle that individuals and organizations are responsible for the outcomes and impacts of AI systems they develop or deploy.

In AI ethics, ensuring that AI systems do not discriminate against certain groups or individuals and provide equitable outcomes.

The ability to understand the data, algorithms, and decision-making processes of an AI system.

The ability of an AI model to maintain its performance and accuracy even when faced with variations, noise, or adversarial attacks in the input data.

Maliciously crafted inputs designed to deceive or cause an AI model to make incorrect predictions.

A model's ability to perform well on new, unseen data, indicating that it has learned general patterns rather than just memorizing the training data.

A hypothetical future point in time when technological growth becomes uncontrollable and irreversible, resulting in unforeseeable changes to human civilization, often associated with the development of ASI.

The study and application of moral principles and values to the design, development, deployment, and use of artificial intelligence systems, ensuring they are beneficial and fair to society.