Linked from

The 10 pages that link to Training data, each with the reason it gives.

• Machine learningRelated: The examples available during training shape what patterns a model can learn.

• Algorithmic biasRelated: Historical gaps or skewed examples can make a model perform unevenly across groups.

• Computer visionRelated: Labels and coverage in training data strongly shape what a vision model can recognize.

• Neural networkRelated: Networks adapt their parameters using patterns present in these examples.

• Deep learningRelated: The examples determine which patterns a deep network can learn.

• Pattern recognitionRelated: Supervised recognizers learn category boundaries from labeled examples.

• Facial recognition systemRelated: Training images shape which facial variations a recognition model learns to handle.

• Large language modelRelated: The text included in training shapes which patterns and information a model can learn.

• AI artRelated: The images and other works in training data influence what a model can generate.

• Generative AIRelated: The examples shape what a generator can reproduce, combine, or omit.