This repository showcases the recovery of quantitative models of human information processing with differentiable architecture search (DARTS).

Requires Python 3.7. Required packages are listed in requirements.txt

We are currently in the process of adding full documentation and tutorials for this code, as part of a pipeline for autonomous empirical research (empiricalresearch.ai). Please stay tuned for updates.

We adopted parts of the code from: https://github.com/quark0/darts

To run the recovery of the weber model with original DARTS

python run_weber_original_study_slurm.py --slurm_id 0

The slurm_id may vary from 0-149 and generates results for different combinations of

• the number of intermediate nodes k and
• the parameter penalty gamma

Liu, H., Simonyan, K., & Yang, Y. (2018). Darts: Differentiable architecture search. arXiv preprint arXiv:1806.09055.

Musslick, S. (2021). Recovering Quantitative Models of Human Information Processing with Differentiable Architecture Search. arXiv preprint arXiv:2103.13939.