AIExplainer
Machine Learning Intermediate

NaN trap

When one number in your model becomes a NaN during training, which causes many or all other numbers in your model to eventually become a NaN.

When one number in your model becomes a NaN during training, which causes many or all other numbers in your model to eventually become a NaN. NaN is an abbreviation for Not a Number.

Practitioners refer to nan trap when building, training, or evaluating machine learning systems. It appears in research papers, product documentation, and technical discussions about AI capabilities and limitations.