← Back to Tutorials

4. AI Terminology

Getting Started

Essential AI Terms

TermDefinition
AgentAn entity that perceives its environment and takes actions to achieve goals
PerceptWhat an agent senses at a given moment from the environment
HeuristicA rule-of-thumb or informed guess that helps solve problems efficiently
OverfittingModel learns training data too well, including noise, and fails on new data
UnderfittingModel is too simple to capture patterns in the data
BiasSystematic error in predictions; also refers to societal bias in AI systems
VarianceSensitivity of model to fluctuations in training data
FeatureAn individual measurable property of a data point used for learning
LabelThe target output in supervised learning (e.g., "cat" or "dog")
TrainingThe process of teaching a model using labeled or unlabeled data
InferenceUsing a trained model to make predictions on new data
Latent SpaceCompressed representation of data learned by a neural network
TokenA unit of text processed by an LLM (word, subword, or character)
AttentionMechanism that lets models focus on relevant parts of input when generating output
Transfer LearningUsing knowledge from one task to improve learning on a related task
Fine-tuningAdapting a pre-trained model for a specific downstream task
Reinforcement LearningLearning through trial-and-error with rewards and penalties
HallucinationWhen an AI generates plausible but incorrect information
Practice Task: Create flashcards for 10 terms from this chapter. For each term, write the definition and a real-world example. Use these terms in a short paragraph describing how an AI recommendation system works.