Hyperparameter: Difference between revisions

m
 
(One intermediate revision by the same user not shown)
Line 29: Line 29:
For example, you could instruct the robot to take one step at a time or two steps at once. This is similar to setting the learning rate hyperparameter on machine learning models; it tells them how much to adjust their predictions based on new data they encounter.
For example, you could instruct the robot to take one step at a time or two steps at once. This is similar to setting the learning rate hyperparameter on machine learning models; it tells them how much to adjust their predictions based on new data they encounter.


You could instruct the robot to pay close attention to other players or focus solely on itself. This is similar to setting a regularization strength hyperparameter in machine learning models - how much attention should be paid to training data versus prevention data.
You could instruct the robot to pay close attention to other players or focus solely on itself. This is similar to setting a regularization strength hyperparameter in machine learning models - how much attention should be paid to training data to prevent overfitting.


==Explain Like I'm 5 (ELI5)==
[[Category:Terms]] [[Category:Machine learning terms]] [[Category:not updated]]
Imagine you want to build a robot that can distinguish between apples and bananas, but aren't sure how. There are some settings you can adjust in order to help the machine improve at distinguishing between different fruits.
 
Hyperparameters are those variables you can alter, like the speed or direction a toy car turns. Hyperparameters include things like robot size, how long it looks at fruits for recognition purposes, and how many pictures need to be shown before recognition improves.
 
By altering these hyperparameters, the robot can become better at recognizing fruit - just like how your toy car can be made faster or slower. So in machine learning, we adjust these hyperparameters in order for the model to perform optimally and become better at what it's supposed to do.
 
 
[[Category:Terms]] [[Category:Machine learning terms]]