Cross-Validation
A model evaluation technique that partitions data into subsets, trains the model on some subsets and validates on others, providing a more reliable estimate of real-world performance.
In plain language
Testing an AI by dividing your data multiple ways and checking how the model performs on each split. Instead of a single test, you run five or ten, getting a stronger signal of how it will work with new, unseen data.
Why this matters
Cross-validation strengthens your confidence in model performance estimates and helps prevent overfitting, which is critical for governance assurance. Your model evaluation processes should require cross-validation rather than single-split testing.
Relevance
ImplementationCross-validation provides stronger confidence in model robustness and helps identify models that look good on paper but fail in the real world.
Related terms
Putting cross-validation into practice in your organisation?
Ready to transform your AI strategy?
Partner with Australia's AI strategy and governance specialists. From adoption roadmaps to ISO 42001 audit readiness.