Inverse Active Sensing
Daniel Jarrett, Mihaela van der Schaar
International Conference on Machine Learning (ICML) 2020
Code Author: Daniel Jarrett (dkj25@cam.ac.uk)
Overview
This repository contains the necessary code to replicate the main experimental results in the ICML 2020 paper "Inverse Active Sensing: Modeling and Understanding Timely Decision-Making." Bayesian Inverse Active Sensing, which is given as an example instance of inverse active sensing in the paper, is implemented in the file learner/bayesian_mcmc_learner.py.
Usage
First, navigate to the directory package/pomdp-sol/pomdp_sol/proc/, and install pomdp-solve v5.4 inside the empty directory nix-solver by following the instructions on pomdp.org. Make sure the executable pomdp-solve is located at nix-solver/src/pomdp-solve. Likewise, install sarsop inside the empty directory nix-sarsop by following the instructions on adacompnus. Make sure the executable pomdpsol is located at nix-sarsop/pomdpsol.
Install the required python packages by running:
python3 -m pip install -r requirements.txt
For generating the figures, make sure you have Latex installed
sudo apt install texlive texlive-latex-extra texlive-fonts-recommended dvipng cm-super
Then, the main experimental results in the paper can be replicated by running:
make pack
make test
make driver
with the appropriate scenarios selected in the file driver.py. The selectable configurations are located in the file config/decision.py and config/diagnosis.py, and can be changed as needed.
Dependencies
The following packages are needed for running the model:
- gym==0.15.3
- matplotlib==3.1.1
- numpy==1.17.3
- pandas==1.1.5
- scikit-learn==0.24.1
- scipy==1.5.2
- torch==1.7.1
Reference
If you use this code, please cite:
@article{jarrett2020inverse,
title = {Inverse Active Sensing: Modeling and Understanding Timely Decision-Making},
author = {Jarrett, Daniel and van der Schaar, Mihaela},
journal = {International Conference on Machine Learning},
year = {2020},
}