Contains experiments conducted and experimental code for MSc dissertation "On the Merits of Biologically Plausible Neural Nets".
The related dissertation can be found here.
/basic-cnn-conversion
: first experiment for CNN-SNN conversion on MNIST using SNN Toolbox & Keras
/lstm-snn-comparison
: time series experiment for SNN conversion, SNN Toolbox & Keras
/bindsnet-snn
: SNN on MNIST using a Diehl & Cook network implemented in BindsNET
/pc-backprop
: contains a comparison of CNN backpropagation training with pred. coding steps with Torch2PC & PyTorch
/pc-reservoir
: contains an (unused) single-module python adaption of (Yonemura & Katori) [1].
/lif-reservoir
: RC with LIF reservoir generation on MNIST + showcase
[1] Y. Yonemura and Y. Katori. Network model of predictive coding based on reservoir computing for multi-modal processing of visual and auditory signals. Nonlinear Theory and Its Applications, IEICE, 12(2):143–156, 2021