Learning Rate and Convergence MCQ | Optimization for AI Multiple Choice Questions

Direction: Choose the correct option

What does the learning rate control in gradient descent?

A. Step size

B. Number of iterations

C. Initial parameters

D. Error tolerance

Direction: Choose the correct option

If learning rate is too high, gradient descent may:

A. Overshoot and diverge

B. Converge slowly

C. Get stuck in local minima

D. Never update

Direction: Choose the correct option

If learning rate is too low, gradient descent will:

A. Converge slowly

B. Overshoot

C. Not update

D. Diverge

Direction: Choose the correct option

Convergence in gradient descent means:

A. Cost function stops decreasing significantly

B. Parameters become zero

C. Gradient becomes infinite

D. Cost function reaches global minimum

Direction: Choose the correct option

What is learning rate scheduling?

A. Adjusting learning rate during training

B. Setting fixed learning rate

C. Calculating initial learning rate

D. None

Quipoin Menu