Key terms
A procedure or set of steps used to solve a problem.
A system trained or configured to produce predictions, classifications, generations, or decisions.
A collection of examples used for training, evaluation, or analysis.
A measurable input used by a machine learning model.
The target answer used in supervised learning.
A numerical representation of meaning used for search, similarity, and retrieval.
A database optimized for storing and searching embeddings.
Retrieval augmented generation: connecting an LLM with external documents or knowledge.
Additional training to adapt a model to a task, style, or domain.
Instructions, context, and examples given to a generative model.
An AI system that uses tools and steps to complete tasks.
Operational practices for deploying, monitoring, and maintaining machine learning systems.
Using a trained model to produce an output.
When a generative model produces confident but incorrect or unsupported content.
Return on investment: the value generated compared with cost.