Lesson 1 of 7
What does 'training' actually mean?
Explanation
A model starts with random or default internal numbers, called parameters (or weights). Training is a loop: 1) make a prediction with the current parameters, 2) measure how wrong it was using a loss function, 3) nudge the parameters slightly in the direction that would have reduced that error, and repeat thousands of times. "Training" is nothing mystical — it is just repeated small corrections driven by a number that measures wrongness.
Visual explanation
Read this from left to right: what goes in, what happens, and what comes out.
The loop repeats many times, making small corrections after each mistake.
In plain language
Training is repeated practice: the model guesses, compares its guess with the right answer, and adjusts itself slightly.
Remember
Training is adjustment through repeated feedback.
What to do
In your own words, what are the three repeated steps inside a training loop?