Methods to train SNN by surrogate gradient and spatial-temporal backpropagation (STBP)? #167
Unanswered
CloudyDory
asked this question in
Q&A
Replies: 1 comment
-
Lava-dl SLAYER is an enhanced version of the original SLAYER paper. It uses surrogate gradient and BPTT similar to STBP you mentioned. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi, is it possible to train SNN by traditional spatial-temporal backpropagation (STBP) and surrogate gradient? For example the method proposed in: Spatio-Temporal Backpropagation for Training High-Performance Spiking Neural Networks? This is implemented in other SNN libraries, such as SpikingJelly and snnTorch. And it seems that the DECOLLE plugin is still different the traditional surrogate gradient approach.
Beta Was this translation helpful? Give feedback.
All reactions