Concept graph
Glossary
Short definitions with deeper context and cross-links to sibling terms.
Neural Networks
Activation Function
A mathematical function that determines the output of a neural network layer.
Machine Learning
Active Learning
A machine learning paradigm where the model can query a user to obtain labels for new data points actively.
signal-processing
adaptive-filtering
A technique for dynamically adjusting filter parameters based on input signal characteristics.
Machine Learning
adaptive-learning
A method where the system optimizes its learning process based on user interactions and performance.
Machine Learning
Adversarial Training
Adversarial training improves model robustness by training on deliberately perturbed inputs that are designed to expose failures under a defined threat model.
Security
adversarial-example
An input designed to fool a machine learning model into making incorrect predictions.
Agents
Agent memory
Agent memory is the state an AI agent keeps across steps or sessions, such as scratchpad notes, retrieved facts, user preferences, or task history.
Agents
Agent2Agent protocol
Agent2Agent, or A2A, is an open protocol from Google for agent-to-agent communication, capability discovery, task management, and artifact exchange.
Agents
Agentic AI
Agentic AI refers to AI systems that can plan, call tools, maintain task state, and take multi-step actions toward a goal.
Agent security
Agentjacking
Agentjacking is an informal term for hijacking an AI agent's tools, context, or execution path so it performs attacker-directed actions.
Safety
AI guardrails
AI guardrails are controls that constrain, monitor, or validate model behavior before outputs or tool actions reach users or systems.
General Concepts
Algorithm
A step-by-step procedure for calculations or problem-solving.
Data Analysis
anomaly
An anomaly is a data point that deviates significantly from the expected pattern.
machine-learning
anomaly-detection
A technique used to identify unusual patterns or outliers in data.
Deep Learning
Artificial Neural Network (ANN)
A computational model inspired by the way biological neural networks function.
Safety
attention
Attention weights how much each token should attend to others when producing representations.
Neural Networks
Attention Mechanism
A technique that allows models to focus on specific parts of the input data when making predictions.
Data Augmentation
augment-data
A technique to artificially increase the size of a training dataset by creating modified copies of existing data.
Data Preparation
Augmented Data
Synthetic data created to enhance the diversity and quantity of training datasets.
Interactive Technology
Augmented Reality
An interactive experience that overlays digital information onto the real world.
deep learning
auto-encoding
A type of artificial neural network used to learn efficient codings of unlabeled data.
neural networks
autoencoder
An autoencoder is a type of neural network used for unsupervised learning of efficient representations.
Machine Learning
Automated Machine Learning (AutoML)
A process that automates the end-to-end process of applying machine learning to real-world problems.
Data Science
automated-data-analysis
The use of algorithms and software to analyze data without human intervention.