AI Foundations
Hyperparameter
A setting chosen before or during training that controls how learning happens.
Example
A team chooses the tree depth before training rather than letting the model learn it.
Why it matters
It lets builders balance learning quality, time and computing needs for the task.
You'll hear it
“Which training settings changed between these two runs?”
Can you control it?
Set by the team
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Can software choose these settings?
- Yes. Automated searches can compare options, although people still choose the task and the limits.
- How is a hyperparameter different from a parameter?
- Training learns parameters, while people or tuning systems choose hyperparameters.
- Is temperature a hyperparameter?
- Often, yes. It usually affects replies rather than how the AI is taught.