Follow an example through the system.
Examples
For a supervised image classifier, people assemble and label example images. Separate evaluation data helps test performance beyond the examples used to fit the model.
Illustrative supervised-learning workflow. AI also includes other learning methods and rule-based systems. No model is running here.
Training versus using AI
Training is the process of adjusting a model from examples. Inference is the name for using a trained model to make a prediction or produce an answer. They are different activities.
A simple example
Imagine showing a system thousands of labelled pictures of different animals. During training it learns visual patterns associated with those labels. Later it may estimate which animal is in a picture it has not seen before.
What about chatbots?
Language models work with pieces of text called tokens. They generate responses using patterns learned during training and the information available in their context. Their answers may sound confident even when they are inaccurate.
CHECK THE ORIGINAL MATERIAL
Sources & further reading
Starting points for the explanation and the illustrative interface above. Sources are external; examples are not CLU capabilities.