How a Model Learns (Training, in Plain English)

What 'training' actually does to a machine learning model - a model is a bundle of adjustable numbers, and training nudges those numbers until its predictions match known examples.

1. Data → Weights → Predictions A model is a big bundle of adjustable numbers called weights; training sets those numbers, and predicting means running a new input through them to get an answer.
2. Learning by Being Wrong Training is a loop: the model predicts, we measure how wrong it was (the loss), we nudge the weights to make it a little less wrong, and we repeat over millions of examples.
3. Overfitting & Why Test Sets Exist A model can memorize its training data instead of learning the real pattern; we hold back data it never sees (validation and test sets) to catch that, and a model is only as fair as the data it learned from.