Model Adaptation · Software component

Active Learning Sampler

Software componentModel AdaptationModelsarc:ActiveLearningSampler

A sampling component that selects production interactions for detailed human review, prioritizing high-uncertainty cases and diverse examples from underrepresented scenarios.

Responsibility. Chooses which conversations warrant human labeling to maximize training value.

Also known as: Active learning

Direct neighbourhood (hover for relationship types)

Relationships

reads dependency

Trace Store Ch10.1

sends data to dynamic

Human Evaluator Ch10.1

Design guidance

SHOULD focus human review on low-confidence cases and diverse examples that expand coverage.

Quantitative guidance

As stated by the sources; verify before use.

Klarna samples 5% of conversations for detailed human review (Ch10.1).

Classification

Patterns: Active learningUncertainty sampling
Quality attributes: Cost efficiencyFunctional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)
Risks mitigated: Wasting reviewer effort on uninformative cases

Sources

Ch10.1: T. Nguyen, "Conversational UI," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 10.1. ISBN: 9798244538229.