Backpropagation through time

Backpropagation through time (BPTT) is a gradient-based technique for training certain types of recurrent neural networks. It can be used to train Elman networks. The algorithm was independently derived by numerous researchers.[1][2][3]

Algorithm edit

 
BPTT unfolds a recurrent neural network through time.

The training data for a recurrent neural network is an ordered sequence of   input-output pairs,  . An initial value must be specified for the hidden state  . Typically, a vector of all zeros is used for this purpose.

BPTT begins by unfolding a recurrent neural network in time. The unfolded network contains   inputs and outputs, but every copy of the network shares the same parameters. Then the backpropagation algorithm is used to find the gradient of the cost with respect to all the network parameters.

Consider an example of a neural network that contains a recurrent layer   and a feedforward layer  . There are different ways to define the training cost, but the aggregated cost is always the average of the costs of each of the time steps. The cost of each time step can be computed separately. The figure above shows how the cost at time   can be computed, by unfolding the recurrent layer   for three time steps and adding the feedforward layer  . Each instance of   in the unfolded network shares the same parameters. Thus the weight updates in each instance ( ) are summed together.

Pseudocode edit

Pseudocode for a truncated version of BPTT, where the training data contains   input-output pairs, but the network is unfolded for   time steps:

Back_Propagation_Through_Time(a, y)   // a[t] is the input at time t. y[t] is the output
    Unfold the network to contain k instances of f
    do until stopping criterion is met:
        x := the zero-magnitude vector // x is the current context
        for t from 0 to n − k do      // t is time. n is the length of the training sequence
            Set the network inputs to x, a[t], a[t+1], ..., a[t+k−1]
            p := forward-propagate the inputs over the whole unfolded network
            e := y[t+k] − p;           // error = target − prediction
            Back-propagate the error, e, back across the whole unfolded network
            Sum the weight changes in the k instances of f together.
            Update all the weights in f and g.
            x := f(x, a[t]);           // compute the context for the next time-step

Advantages edit

BPTT tends to be significantly faster for training recurrent neural networks than general-purpose optimization techniques such as evolutionary optimization.[4]

Disadvantages edit

BPTT has difficulty with local optima. With recurrent neural networks, local optima are a much more significant problem than with feed-forward neural networks.[5] The recurrent feedback in such networks tends to create chaotic responses in the error surface which cause local optima to occur frequently, and in poor locations on the error surface.

See also edit

References edit

  1. ^ Mozer, M. C. (1995). "A Focused Backpropagation Algorithm for Temporal Pattern Recognition". In Chauvin, Y.; Rumelhart, D. (eds.). Backpropagation: Theory, architectures, and applications. Hillsdale, NJ: Lawrence Erlbaum Associates. pp. 137–169. Retrieved 2017-08-21. {{cite book}}: |website= ignored (help)
  2. ^ Robinson, A. J. & Fallside, F. (1987). The utility driven dynamic error propagation network (Technical report). Cambridge University, Engineering Department. CUED/F-INFENG/TR.1.
  3. ^ Werbos, Paul J. (1988). "Generalization of backpropagation with application to a recurrent gas market model". Neural Networks. 1 (4): 339–356. doi:10.1016/0893-6080(88)90007-x.
  4. ^ Sjöberg, Jonas; Zhang, Qinghua; Ljung, Lennart; Benveniste, Albert; Delyon, Bernard; Glorennec, Pierre-Yves; Hjalmarsson, Håkan; Juditsky, Anatoli (1995). "Nonlinear black-box modeling in system identification: a unified overview". Automatica. 31 (12): 1691–1724. CiteSeerX 10.1.1.27.81. doi:10.1016/0005-1098(95)00120-8.
  5. ^ M.P. Cuéllar and M. Delgado and M.C. Pegalajar (2006). "An Application of Non-Linear Programming to Train Recurrent Neural Networks in Time Series Prediction Problems". Enterprise Information Systems VII. Springer Netherlands. pp. 95–102. doi:10.1007/978-1-4020-5347-4_11. ISBN 978-1-4020-5323-8.