Skip to content

Presentation #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 4 commits into from
Apr 6, 2023
Merged

Presentation #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
274108b
notebook for presentation
Z80coder Apr 5, 2023
7d98a96
update readme
Z80coder Apr 6, 2023
c5bcc1c
Merge branch 'main' into presentation
Z80coder Apr 6, 2023
cab9a91
Update test_mnist.py
Z80coder Apr 6, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 16 additions & 6 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,20 +1,30 @@
# dB nets
# ∂B nets

[![Python package](https://github.com/Z80coder/discrete-differentiable-networks/actions/workflows/python.yaml/badge.svg)](https://github.com/Z80coder/discrete-differentiable-networks/actions/workflows/python.yaml)

A Boolean Logic Network library for learning boolean functions on GPUs with gradient descent.
A neural network library for learning boolean-valued, discrete functions on GPUs with gradient descent.

The working prototype is implemented in Wolfram. The production library is implemented in Python.
The library is implemented in Python using the [Flax](https://github.com/google/flax) and [JAX](https://github.com/google/jax) frameworks.

Questions? Ask @Z80coder

## Draft paper

["∂B nets: learning discrete functions by gradient descent"](./docs/db.pdf) (March 2023).

## Demos

Neural logic nets
[∂B nets](https://drive.google.com/file/d/1UUhv6loBrFnZ7jwiHBofnp06at_8bm_F/view?usp=share_link) (1 hour).

# Prototype

The working prototype was implemented in Wolfram. The demos below were snapshots of work-in-progress.

## Demos

- [Neural logic nets](https://drive.google.com/file/d/1_IECuI0f58o_aIIdaQhRo6qPH517YaMa/view?usp=share_link) (15m)

## Development videos
## Development snapshots

- [The Soft-NOT operator](https://drive.google.com/file/d/1z2WFpz4eWLb9xauRnIl6mSXhkbU-XR6X/view?usp=share_link) (10m)
- [The Soft-AND operator](https://drive.google.com/file/d/1l9Y2cWJYYdYSsgqwfH-Dfo2Nxmiewia-/view?usp=share_link) (10m)
Expand All @@ -33,4 +43,4 @@ Neural logic nets
- [Boolean logic nets and MNIST](https://drive.google.com/file/d/12Rwx8H76UTNRdBK4WAwe_QeTWiGrbP-_/view?usp=share_link) (18m)
- [Neural logic nets for differentiable QL](https://drive.google.com/file/d/15rAagCh7LxEN0CHVNkTY6iPWSxrAG0pW/view?usp=share_link) (30m)

More to come!

Loading