Translations:Gradient Descent/16/en
| Variant | Gradient computed over | Per-step cost | Gradient noise |
|---|---|---|---|
| Batch (full) gradient descent | All $ N $ samples | High | None |
| stochastic gradient descent (SGD) | 1 random sample | Low | High |
| mini-batch gradient descent | $ B $ random samples ($ 1 < B < N $) | Medium | Medium |