AI Foundations
Sigmoid function
An S-shaped calculation often used to turn scores into numbers between zero and one.
Example
A prediction tool turns a raw score into a number between zero and one.
Why people use it
It turns a wide range of scores into values between zero and one.
What you'll hear
“Map this score onto a bounded scale.”
What this means for you
Compare the stated chances with what actually happens.
Can you control it?
Developer-only
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 it turn a very large score into exactly the same-sized chance?
- No. Its result approaches an upper limit rather than growing without bound.
- Can very large scores produce only tiny changes afterward?
- Yes. Near either end of the curve, further score changes have a smaller effect.
- Is it used only for final predictions?
- No. Related bounded calculations can also be used inside a network.