Skip to content

Latest commit

 

History

History
16 lines (11 loc) · 639 Bytes

README.md

File metadata and controls

16 lines (11 loc) · 639 Bytes

Biological_Learning

Example of "biological" learning for MNIST based on the paper Unsupervised Learning by Competing Hidden Units by D.Krotov and J.Hopfield. If you want to learn more about this work you can also check out this lecture from MIT's 6.S191 course.

Getting started

install jupyter notebook and numpy, scipy, matplotlib.

> jupyter notebook

run Unsupervised_learning_algorithm_MNIST.ipynb and observe weights.

Author and License

(c) 2018 Dmitry Krotov -- Apache 2.0 License