Artificial Intelligence Glossary

A mathematical function applied to a neuron's output that introduces non-linearity into the network, enabling it to learn complex patterns. Common examples include ReLU, sigmoid, and tanh.

Systematic and unfair discrimination in AI outputs that arises from biased training data, flawed design choices, or unrepresentative datasets, leading to inequitable outcomes.

A hypothetical form of AI that would possess human-level cognitive abilities across all intellectual tasks, including reasoning, learning, and creativity. AGI does not yet exist.

A technique in neural networks that dynamically assigns different weights to different parts of the input, allowing the model to focus on the most relevant information for each output step.

An algorithm for training neural networks that computes gradients of the loss function with respect to each weight by propagating errors backward through the network layers.

An AI application that simulates human conversation through text or voice, using NLP techniques to understand user queries and generate appropriate responses.

A supervised learning task where the model assigns input data to one of several predefined categories, such as classifying emails as spam or not spam.

An unsupervised learning technique that groups similar data points together based on shared characteristics, without requiring predefined labels.

The field of AI that enables machines to interpret and understand visual information from images and video, including object detection, recognition, and scene analysis.

A deep learning architecture designed for processing structured grid data like images, using convolutional filters to automatically detect spatial hierarchies of features.

A subset of machine learning that uses neural networks with multiple hidden layers to learn hierarchical data representations, enabling breakthroughs in vision, language, and speech.

A rule-based AI program that emulates the decision-making ability of a human expert within a specific domain by encoding domain knowledge as if-then rules.

AI systems capable of creating new content such as text, images, audio, or video by learning statistical patterns from training data and producing novel outputs.

An iterative optimization algorithm that adjusts model parameters in the direction that minimizes the loss function, forming the basis of most neural network training.

A deep learning model trained on vast text datasets that can understand and generate human language, typically built on the transformer architecture with billions of parameters.

A branch of AI where systems learn to perform tasks by identifying patterns in data and improving through experience, rather than following explicitly coded instructions.

The branch of AI focused on enabling computers to understand, interpret, and generate human language in both text and speech form.

A computational model composed of layers of interconnected nodes (neurons) with adjustable weights, capable of learning complex patterns from data through training.

A condition where a machine learning model learns the training data too closely, including noise and outliers, resulting in poor performance on new unseen data.

A neural network architecture designed for sequential data where connections between nodes form cycles, allowing the network to maintain a form of memory across time steps.

A machine learning paradigm where an agent learns to make decisions by receiving rewards or penalties from an environment, optimizing for maximum cumulative reward over time.

A machine learning approach in which a model is trained on labeled input-output pairs, learning to generalize the mapping from inputs to correct outputs.

A technique where a pre-trained model is adapted for a new but related task, leveraging knowledge learned from one domain to improve performance in another.

A neural network architecture based on self-attention mechanisms that processes input data in parallel, serving as the foundation for modern large language models and many vision models.

A machine learning approach where models learn from unlabeled data, discovering hidden patterns, structures, or groupings without predefined output categories.