Skip to content
Learning Steady-States of Iterative Algorithms over Graphs
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
graphnn @ bdb4e68

graph steady state embedding

Learning Steady-States of Iterative Algorithms over Graphs (

1. Setup the environment

1) Download the repository
git clone --recursive
2) Build the dependency

This project depends on the graphnn library. The building instruction can be found here:
3) Download the data

Use the following dropbox link:

Put everything under the 'data' folder, or create a symbolic link with name 'data':

ln -s /path/to/your/downloaded/files data

Finally the folder structure should look like this:

steady_state_embedding (project root)
|__  code
|__  graphnn
|__  data
|__  |__ algo_data
|    |__ amazon
|    |__ pagerank_ba

2. Run the experiments

Most of the experiments are self-contained, where you need to build the code and then execute the script. For example:

cd code/fit_algo/connectivity

The data is already cooked, so the code/data_process folder is for reference purpose.


  title={Learning Steady-States of Iterative Algorithms over Graphs},
  author={Dai, Hanjun and Kozareva, Zornitsa and Dai, Bo and Smola, Alex and Song, Le},
  booktitle={International Conference on Machine Learning},
You can’t perform that action at this time.