StackedML
Practice
Labs
Questions
Models
Pricing
Sign in
Questions
/
Optimization
/
Gradient Methods
/
Learning rate effects
← Previous
Next →
480.
Low Learning Rate Effect
easy
What happens when the learning rate is set too low in gradient descent?
A
Training produces an overfit model because small steps allow the optimizer to memorize individual training samples
B
Training converges to a suboptimal local minimum since small steps cannot escape early flat regions
C
Training converges very slowly, requiring many more iterations to reach an acceptable minimum
D
Training diverges because small learning rates cause gradient magnitudes to grow exponentially over iterations
Sign in to verify your answer
← Back to Questions