Mon Prof Augmenté
Learning from many photos
Open to everyone until Tuesday, September 29
To keep it, print it while it is open.
A program can get better at telling cats from dogs by looking at many labelled photos. It adjusts itself from those examples instead of following only rules written by hand.
In the picture
- A square photo card on the belt: one example cat or dog
- The dark computer housing with a lit slot: where the examples are read in
- The hovering square photo: a new picture, never seen before
- The bright outline around the animal: the answer the program chose
Examples
- A spam filter learning from the emails you marked as junk.
- A phone keyboard learning the words you type most often.
Key takeaways
- Training on many examples lets a program handle cases nobody wrote rules for.
- More varied examples usually mean better answers on new pictures.
Carry on in the app
Three questions on this lesson, to check the idea stuck. Free in the app while the lesson is in the free week.
At least one new illustrated lesson every weekday for your child’s year, with its quiz and a reminder at 6:15 pm.
Get the app
Free app, optional subscription
App Store Google Play
