AI Foundations
Optimization
The process of changing model parameters to reduce error or improve an objective.
Example
The training process repeatedly changes weights to reduce prediction error.
Why it matters
It helps search through many possible internal settings without trying every combination by hand.
You'll hear it
“The score has stopped improving even though training is still running.”
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 different starting points lead to different results?
- Yes. The search can follow different paths and settle on different solutions.
- Does optimization find the perfect model?
- No. The result depends on the examples, method and available computing time.
- How is optimization different from training?
- Training includes preparing and learning from examples; optimization is the part that adjusts values toward the chosen target.