ConceptBatch Gradient Descent / BGD
Batch Gradient Descent
A variant of gradient descent in which each iteration uses all training samples to compute the average gradient and update the weights once; the update direction is stable, but computational cost grows linearly with the number of samples.
Timeline (last 90 days)
Oct 19
The training process uses batch gradient descent for 5000 iterations.
Unverified50%
Sep 11
Batch gradient descent uses all data to compute the gradient each time, resulting in a stable direction but high computational cost.
Unverified50%