greydanus / np_nets

Neural network experiments written purely in numpy

np_nets

Neural network experiments written purely in numpy

1. Learning backprop with the MNIST classification task

• Jupyter notebook

2. Better version of learning backprop...lacks markdown explanations...doesn't have a bug though

• Jupyter notebook

3. Synthetic gradients with the MNIST classification task

• Jupyter notebook
• also check out the minimalist 145-line Gist for this project
• inspired by this Google DeepMind paper

4. Hebbian learning (for a Dartmouth class)

• Jupyter notebook
• For Human Memory (PSYC 051.09) taught by Jeremy Manning

5. "U loss" learning

• I test an ansatz for layer-wise training of neural networks. It didn't work. That's how research goes.
• Folder is here