Backpropagation through structure

Backpropagation Through Structure (BPTS) is a gradient-based technique for training Recursive Neural Nets (a superset of Recurrent Neural Nets) and is extensively described in a 1996 paper written by Christoph Goller and Andreas Küchler.[1]

References

  1. ↑ Kuchler, Andreas. "Learning Task-Dependent Distributed Representations by Backpropagation Through Structure". psu.edu. psu.edu. Retrieved 20 April 2015.