https://github.com/networks-learning/counterfactual-explanations-mdp

Code for "Counterfactual Explanations in Sequential Decision Making Under Uncertainty", NeurIPS 2021
https://github.com/networks-learning/counterfactual-explanations-mdp

cognitive-therapy counterfactual-explanations counterfactuals explainable-ai explainable-ml machine-learning markov-deci mdp sequential-decision-making-problems

Last synced: about 5 hours ago
JSON representation

Code for "Counterfactual Explanations in Sequential Decision Making Under Uncertainty", NeurIPS 2021

• Host: GitHub
• URL: https://github.com/networks-learning/counterfactual-explanations-mdp
• Owner: Networks-Learning
• Created: 2021-10-28T16:34:21.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2023-02-08T12:35:32.000Z (almost 2 years ago)
• Last Synced: 2024-04-16T02:14:54.136Z (7 months ago)
• Topics: cognitive-therapy, counterfactual-explanations, counterfactuals, explainable-ai, explainable-ml, machine-learning, markov-deci, mdp, sequential-decision-making-problems
• Language: Jupyter Notebook
• Homepage:
• Size: 120 KB
• Stars: 13
• Watchers: 2
• Forks: 3
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Counterfactual Explanations in Sequential Decision Making Under Uncertainty

This repository contains the code used in the paper [*Counterfactual Explanations in Sequential Decision Making Under Uncertainty*](https://arxiv.org/abs/2107.02776), published at NeurIPS 2021.

## Dependencies

All the experiments were performed using Python 3.9. In order to create a virtual environment and install the project dependencies you can run the following commands:

```bash

python3 -m venv env

source env/bin/activate

pip install -r requirements.txt

```

## Code organization

The directory `src` contains the source code for the various experiments. The following table contains a short description for each python file:

| `src/`                                                   | Description                                                  |

| ------------------------------------------------------------ | ------------------------------------------------------------ |

| [synth_mdp.py](src/synth_mdp.py)                           | Defines a **Synth_MDP** object which handles the generation of synthetic decision making realizations and their counterfactual explanations. |

| [synth_compute_cf_mdps.py](src/synth_compute_cf_mdps.py)    | Generates synthetic decision making realizations, pre-computes the counterfactual transition probabilties for each one of them and saves them as `.pkl` files. |

| [synth_experiment.py](src/synth_experiment.py)                     | Generates optimal counterfactual explanations for each synthetic decision making realization, for various values of the parameter `k`. |

| [therapy_mdp.py](src/therapy_mdp.py)                           | Defines a **Therapy_MDP** object which handles the input of the cognitive behavioral therapy data and the generation of counterfactual explanations for each patient. |

| [therapy_compute_cf_mdps.py](src/therapy_compute_cf_mdps.py)    | Pre-computes the counterfactual transition probabilties for all patients' realizations and saves them as `.pkl` files. |

| [therapy_experiment.py](src/therapy_experiment.py)                     | Generates optimal counterfactual explanations for each patient's realization, for various values of the parameter k. |

| [therapy_evaluation.py](src/therapy_experiment.py)                     | Generates counterfactual explanations for each patient's realization, using our method and various baselines. |

The directory [scripts](scripts/) contains bash scripts that use the aforementioned code and pass parameter values required for the various experiments.

The directory [notebooks](notebooks/) contains jupyter notebooks producing the figures appearing in the paper. Each notebook has script execution prerequisites specified therein.

The directory [figures](figures/) is used for saving the figures produced by the notebooks.

The directory [outputs](outputs/) is used for saving the json outputs produced by the scripts. The sub-directory [cf_mdps](outputs/cf_mdps/) is used for saving intermediate `.pkl` files containing the counterfactual transition probabilities for each single realization.

## Citation

If you use parts of the code in this repository for your own research, please consider citing:

    @inproceedings{tsirtsis2021counterfactual,

        title={Counterfactual Explanations in Sequential Decision Making Under Uncertainty},

        author={Tsirtsis, Stratis and De, Abir and Gomez-Rodriguez, Manuel},

        booktitle={Advances in Neural Information Processing Systems},

        year={2021}

    }