sample efficiency
Sample efficiency measures how much useful learning an AI system achieves from a given number of training examples or experiences.
AI Terminology
Search clear AI terminology definitions, acronyms, related concepts and the reviewed videos that explain them.
Showing 401–420 of 1136 terms
Clear filtersself-service agent
A self-service AI agent lets users request and use capabilities without requiring a specialist to configure every workflow manually.
small language model
SLMA small language model is a language model with a comparatively compact architecture or resource footprint, designed to run efficiently on bounded tasks and hardware.
stablecoin
A stablecoin is a digital token designed to maintain a relatively stable value by referencing an external asset such as a national currency.
supercomputing provider
An AI supercomputing provider operates large accelerator clusters and sells managed access to organizations training or serving AI models.
synthetic data
Synthetic data is generated rather than directly observed data, used for purposes such as training or testing AI systems.
terminal multiplexer
A terminal multiplexer manages multiple persistent terminal sessions, windows or panes through one interface and can keep them running when the interface disconnects.
test-driven development
TDDTest-driven development is a software workflow in which a failing test is written before implementation, then code is added and refactored until the test passes.
text watermarking
Text watermarking places a hidden statistical pattern in AI-generated wording or formatting so a compatible detector can find evidence of model involvement.
trading bot
An AI trading bot uses AI-assisted analysis or decision logic to generate and execute trades under defined rules and controls.
unattended AI agent
An unattended AI agent performs background work without a person supervising every execution step.
unreleased model
An unreleased model is an AI model developed or trained internally but not made available for public use or independent access.
vector search
Vector search finds items by comparing numerical representations of their features or meaning.
vertical integration
Vertical integration brings multiple layers of an AI product stack, such as models, runtimes, tools, and distribution, under one organization's control.
virtual private server
A virtual private server is an isolated software-defined server environment that runs on shared physical hardware and can host applications continuously.
watermark robustness
Watermark robustness is the degree to which a hidden content mark remains detectable after editing, rewriting, formatting changes, compression, or other transformations.
WebSocket
A WebSocket is a persistent two-way network connection that lets a client and server exchange messages with low overhead.
4D reconstruction
4D reconstruction estimates a three-dimensional subject or scene and how its shape and position change over time.
action-item extraction
Action-item extraction identifies tasks, owners, deadlines, and commitments in conversations, documents, or meeting transcripts.
activation classifier
An activation classifier analyzes internal AI model activations to detect patterns associated with selected concepts or behaviors.