CB-SOM

Credit Based Self Organization for Learning Deep Topographically Organized Representations

We are actively updating the repository to replicate the plots in the paper.

To run the notebook, you need to install the TDANN package. You can do so by cloning the repository from the following link:

https://github.com/neuroailab/TDANN.git

Installation Instructions

1. Clone the repository:

   git clone https://github.com/neuroailab/TDANN.git

2. Navigate to the directory:

   cd TDANN

3. Install dependencies:

   pip install -r requirements.txt

4. Install the package:

   python setup.py install

You can find the checkpoints here https://drive.google.com/drive/folders/1XYs1bWB17ycuyXrl67ACwD2X0Ar1xlVH?usp=share_link

📜 Citation:

If you use our work, please cite:

@inproceedings{
dehghani2025creditbased,
title={Credit-based self organizing maps: training deep topographic networks with minimal performance degradation},
author={Amirozhan Dehghani and Xinyu Qian and Asa Farahani and Pouya Bashivan},
booktitle={The Thirteenth International Conference on Learning Representations},
year={2025},
url={https://openreview.net/forum?id=wMgr7wBuUo}
}