Feedback Alignment

• Tensorflow implementation of Direct feedback Alignment Provides Learning in Deep Neural Networks and Random synaptic feedback weights support error backpropagation for deep learning.

• DataDownload.ipynb is for downloading notMNIST data. This code is from the Tensorflow Udacity source code

• DFA.ipynb is the implementation of Direct feedback Alignment.

• RFA.ipynb is the implementation of Random feedback Alignment.