backpropagation through time
"Backpropagation through time" (BPTT) is a technique used in recurrent neural networks (RNNs) to train the network by recursively applying the backpropagation algorithm to unfold the network over a sequence of time steps. It enables the network to learn the temporal dependencies in sequential data by updating the weights based on the error signals propagated backwards in time through the unfolded network structure.
Requires login.
Related Concepts (1)
Similar Concepts
- backpropagation through time (bptt)
- backward chaining
- backward pass
- batch gradient descent
- error backpropagation
- forward pass
- inference and forward propagation
- retrocausality
- temporal causality
- time loops
- time series regression
- time travel
- time travel and causal loops
- time travel and determinism
- time travel paradoxes