AI_GLOSSARY.exe

The percentage of correct predictions made by a model, calculated as the number of correct predictions divided by the total number of predictions.

A mathematical function that enables neural networks to learn complex, nonlinear relationships between features and outputs. Common examples include ReLU and Sigmoid.

An AI system that can autonomously perceive its environment, make decisions, and take actions to achieve specific goals.

A theoretical form of AI that can perform any intellectual task that a human can do, with the ability to learn and adapt across multiple domains.

A set of step-by-step instructions that allows a computer program to learn from data, recognize patterns, and accomplish tasks autonomously.

The tendency to attribute human characteristics, emotions, or consciousness to AI systems, such as believing a chatbot has feelings.

A set of rules and protocols that allows different software applications to communicate with each other, commonly used to access AI services.

The simulation of human intelligence in machines that are programmed to think, learn, and solve problems like humans.

A technique in neural networks that allows the model to focus on specific parts of the input when making predictions, weighing their importance dynamically.

A neural network that learns to compress data into a lower-dimensional representation and then reconstruct it, useful for dimensionality reduction and anomaly detection.

The use of AI and technology to perform tasks without human intervention, increasing efficiency and reducing manual labor.

A fundamental algorithm for training neural networks that calculates gradients and propagates errors backward through the network to update weights.

The number of training examples processed together in one iteration before the model updates its parameters.

Systematic errors in AI models resulting from flawed training data, leading to unfair or inaccurate predictions for certain groups or scenarios.

Extremely large datasets that are too complex for traditional data processing methods, often used to train AI models.

A type of classification task where the model predicts one of two possible outcomes, such as yes/no or true/false.

An AI program that simulates human conversation through text or voice, designed to interact with users and provide information or assistance.

A machine learning task where the model assigns input data to predefined categories or classes.

The delivery of computing services including AI processing power, storage, and databases over the internet.

An unsupervised learning technique that groups similar data points together based on their characteristics.

A specialized neural network architecture designed for processing grid-like data such as images, using convolutional layers to detect patterns and features.

A field of AI that enables machines to interpret and understand visual information from the world, such as images and videos.

The process of discovering patterns, correlations, and useful information from large datasets using AI and statistical methods.

The process of cleaning, transforming, and organizing raw data before feeding it into a machine learning model.

A collection of data used to train, validate, and test AI models, typically consisting of examples with inputs and corresponding outputs.

A subset of machine learning that uses multi-layered neural networks to learn complex patterns from large amounts of data.

A generative AI technique that learns to create data by reversing a process that gradually adds noise to training examples.

A numerical representation of data (such as words or images) in a lower-dimensional space that captures semantic relationships.

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

Principles and guidelines aimed at ensuring AI systems are developed and used responsibly, fairly, and without causing harm to individuals or society.

The ability to understand and interpret how an AI model makes its decisions, important for trust and accountability.

An individual measurable property or characteristic of data used as input for machine learning models.

The process of selecting, creating, and transforming features from raw data to improve model performance.

The process of taking a pre-trained model and further training it on a specific dataset to adapt it for a particular task or domain.

A type of AI architecture consisting of two neural networks—a generator and discriminator—that compete to create realistic synthetic data.

AI systems capable of creating new content such as text, images, music, or code based on patterns learned from training data.

A specialized processor originally designed for graphics but now widely used to accelerate AI training and inference due to parallel processing capabilities.

An optimization algorithm that iteratively adjusts model parameters to minimize the loss function by moving in the direction of steepest descent.

Safety measures and restrictions implemented in AI systems to prevent harmful, biased, or inappropriate outputs.

When an AI model generates false or nonsensical information while presenting it confidently as if it were factual.

A configuration setting that controls the learning process of a model, such as learning rate or batch size, set before training begins.

The ability of AI systems to identify objects, people, places, and actions in images, a key application of computer vision.

The process of using a trained AI model to make predictions or generate outputs on new, unseen data.

A computer vision task that identifies and delineates each individual object instance in an image at the pixel level.

The correct answer or target output associated with a training example in supervised learning.

The time delay between when an AI system receives an input and when it produces an output or response.

A hyperparameter that controls how much model weights are adjusted during training in response to the calculated error.

A type of AI model trained on massive amounts of text data to understand and generate human-like language.

A mathematical function that measures how well a model's predictions match the actual target values, guiding the optimization process.

A branch of AI that enables computers to learn from data and improve their performance without being explicitly programmed for every task.

A mathematical representation of a system or process that has been trained on data to make predictions or decisions.

AI systems that can process and integrate multiple types of input data, such as text, images, audio, and video simultaneously.

A computational model inspired by the human brain, consisting of interconnected nodes (neurons) organized in layers that process information.

A field of AI focused on enabling computers to understand, interpret, and generate human language in a meaningful way.

A data preprocessing technique that scales features to a similar range, improving model training stability and performance.

A computer vision task that identifies and locates objects within images or videos, often drawing bounding boxes around them.

When a model learns the training data too well, including noise and outliers, resulting in poor performance on new, unseen data.

Internal variables of a model that are learned from training data and determine how the model transforms inputs into outputs.

The ability of AI systems to identify regularities, trends, and structures in data automatically.

A model that has already been trained on a large dataset and can be used as a starting point for new tasks, saving time and resources.

A metric measuring the proportion of positive predictions that are actually correct, calculated as true positives divided by all positive predictions.

The output generated by a trained AI model when given new input data, representing the model's best estimate or decision.

The input text or instruction given to an AI model, especially language models, to elicit a specific response or output.

The practice of carefully crafting input prompts to optimize the quality and relevance of outputs from AI language models.

A metric measuring the proportion of actual positive cases that are correctly identified, calculated as true positives divided by all actual positives.

An AI system that suggests products, content, or services to users based on their preferences, behavior, and similar users' patterns.

A machine learning task where the model predicts continuous numerical values rather than categories.

A type of machine learning where an agent learns to make decisions by taking actions in an environment and receiving rewards or penalties.

A neural network architecture designed to process sequential data by maintaining memory of previous inputs through recurrent connections.

The field combining AI with mechanical engineering to create intelligent machines that can perform physical tasks autonomously.

A computer vision task that classifies every pixel in an image into predefined categories, creating a detailed scene understanding.

An NLP technique that determines the emotional tone or opinion expressed in text, such as positive, negative, or neutral.

The ability of AI systems to convert spoken language into text, enabling voice-based interfaces and commands.

A machine learning approach where models are trained on labeled data, learning to map inputs to known correct outputs.

Artificially generated data created by AI systems rather than collected from real-world events, used for training and testing models.

A parameter that controls the randomness of AI model outputs, with higher values producing more creative and varied responses.

A portion of data held back from training and used only at the end to evaluate the final performance of a model.

The ability of AI models to create human-like written content, from simple sentences to complex articles and stories.

A basic unit of text processed by language models, typically representing a word, part of a word, or punctuation mark.

The process of teaching a machine learning model by exposing it to data and adjusting its parameters to minimize prediction errors.

The portion of data used to teach a machine learning model by adjusting its parameters based on examples.

A technique where a model trained on one task is adapted and fine-tuned for a different but related task, saving time and computational resources.

A neural network architecture that uses self-attention mechanisms to process sequential data in parallel, forming the basis of modern LLMs.

A test of a machine's ability to exhibit intelligent behavior indistinguishable from that of a human, proposed by Alan Turing.

When a model is too simple to capture the underlying patterns in the training data, resulting in poor performance on both training and test data.

A machine learning approach where models find patterns and structures in unlabeled data without explicit guidance on what to learn.

A portion of data held out during training to evaluate model performance and tune hyperparameters without touching the test set.

An AI-powered software agent that can perform tasks or services for users based on voice commands or text input, like Siri or Alexa.

A neural network architecture that applies transformer models to computer vision tasks, treating images as sequences of patches.

A numerical parameter in a neural network that determines the strength of connections between neurons and is adjusted during training.

The ability of a model to perform tasks or recognize categories it has never been explicitly trained on, using only general knowledge.