Trained learned planners
This repository contains the trained networks from the paper "Planning behavior in a recurrent neural network that plays Sokoban", presented at the ICML 2024 Mechanistic Interpretability Workshop.
To load and use the NNs, please refer to the learned-planner repository, and possibly to the training code .
Model details
Hyperparameters:
See model/*/cp_*/cfg.json
for the hyperparameters that were used to train a particular run.
Best Models:
The best models for each of the model type are stored in the following directory:
- DRC(3, 3):
drc33/bkynosqi/cp_2002944000
- DRC(1, 1):
drc11/eue6pax7/cp_2002944000
- ResNet:
resnet/syb50iz7/cp_2002944000
Parameter counts:
- DRC(3, 3): 1,285,125 (1.29M)
- DRC(1, 1): 987,525 (0.99M)
- ResNet: 3,068,421 (3.07M)
Training dataset:
The Boxoban set of levels by DeepMind.
Citation
If you use these neural networks, please cite our work:
@inproceedings{garriga-alonso2024planning,
title={Planning behavior in a recurrent neural network that plays Sokoban},
author={Adri{\`a} Garriga-Alonso and Mohammad Taufeeque and Adam Gleave},
booktitle={ICML 2024 Workshop on Mechanistic Interpretability},
year={2024},
url={https://openreview.net/forum?id=T9sB3S2hok}
}