- Type Parameters:
V - The value type parameter of the tensor whose gradients are being optimized.
- All Implemented Interfaces:
- Component<Tensor<V>>, Optimization<V>, Optimizer<V>
public class SGD<V>
extends java.lang.Object
implements Optimizer<V>
Stochastic Gradient Descent is an iterative optimization technique
that uses the gradient of a weight variable to adjust said variable,
in order to reduce the error used to calculate said gradient.
This simple stochastic gradient descent algorithm does not
optimize the gradient based on previous gradients (network forward and backward passes)
but simply applies the gradient value based on each example within
the training dataset.