# Inferring Human Intentions from Predicted Action Probabilities

*Lei Shi, Paul Bürkner, Andreas Bulling*

*University of Stuttgart, Stuttgart, Germany*

Accepted by [Workshop on Theory of Mind in Human-AI Interaction at CHI 2024](https://theoryofmindinhaichi2024.wordpress.com/)

## Requirements
The code is test in Ubuntu 20.04.

```
pytorch 1.11.0
matplotlib 3.3.2
pickle 4.0
pandas 1.4.3
R 4.2.1
RStan 2.26.3
```
To install R, [see here](https://cran.r-project.org/bin/linux/ubuntu/fullREADME.html)

To install RStan, [see here](https://mc-stan.org/users/interfaces/rstan.html)

## Experiments

To train the neural network model and evaluate the method on Watch-And-Help dataset, see [here](watch_and_help/README.md)

To train the neural network model and evaluate the method on Keyboard and Mouse Interaction dataset, see [here](keyboard_and_mouse/README.md)