Q. The value of the learning rate parameter in a Multi-layer Feed Forward Neural Network(MLFFNN) ranges between :
Solution:
-
If the learning rate is low, then training is more reliable, but optimization will take a lot of time because steps towards the minimum of the loss function are tiny.
-
If the learning rate is high, then training may not converge or even diverge. Weight changes can be so big that the optimizer overshoots the minimum and makes the loss worse
-
We might start with a large value like 0.1, then try exponentially lower values: 0.01, 0.001, etc.
Related Questions
Q. A device used to detect heat radiation is known as:
View SolutionQ. The input impedance of a JFET is high because __—
View SolutionQ. Full form of SMPS IS :
View Solution
Get Question Bank
Strengthen Your Practice with our comprehensive question bank.
