Backpropagation through structure
Jump to navigation
Jump to search
 | This article needs additional citations for verification. (May 2015) |
Backpropagation through structure (BPTS) is a gradient-based technique for training recursive neural networks, proposed in a 1996 paper written by Christoph Goller and Andreas Küchler.[1]
References
[edit | edit source]- ^ Lua error in Module:Citation/CS1/Configuration at line 2172: attempt to index field '?' (a nil value).
 | This artificial neural network-related article is a stub. You can help Wikipedia by expanding it. |