Model Training & Adaptation
Continual learning
Updating a model over time while trying to preserve previously learned capabilities.
Example
A service model keeps learning from new cases after its initial release.
Why it matters
It helps a system stay useful as new examples and situations appear.
You'll hear it
“It learned the new cases, but did the older tasks get worse?”
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
- Does continual mean uninterrupted learning every second?
- No. Updates may happen in planned stages or after enough new material is collected.
- Can new examples be misleading?
- Yes. Poor-quality or unrepresentative additions can steer behavior in the wrong direction.
- Can retaining earlier examples help?
- It can. Revisiting earlier material is one way to reduce loss of older abilities.