Batch Gradient Descent
/bætʃ ˈɡreɪdiənt dɪˈsɛnt/
A gradient descent variant that computes gradients using the entire training dataset in each iteration.
In italiano: Batch Gradient DescentBatch GD provides stable, accurate gradient estimates but is computationally expensive for large datasets. It guarantees convergence to global minimum for convex problems.
Examples
- Full batch training on small datasets
- Theoretical analysis
- Deterministic optimization