Hands-On Machine Learning for Algorithmic Trading

Backpropagation through time

The unrolled computational graph shown in the preceding diagram highlights that the learning process necessarily encompasses all time steps included in a given input sequence. The backpropagation algorithm, which updates the weight parameters based on the gradient of the loss function with respect to the parameters, involves a forward pass from left to right along the unrolled computational graph, followed by a backward pass in the opposite direction.

Just like the backpropagation techniques discussed in Chapter 16, Deep Learning, the algorithm evaluates a loss function to obtain the gradients and update the weight parameters. In the RNN context, backpropagation runs from right to left in the computational graph, ...