Skip to content

Codebase supporting NeurIPS 2024 Paper: Designs for Enabling Collaboration in Human-Machine Teaming via Interactive and Explainable Systems

Notifications You must be signed in to change notification settings

CORE-Robotics-Lab/Team-Development-with-Transparent-Policies

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Designs for Enabling Collaboration in Human-Machine Teaming via Interactive and Explainable Systems

This is the codebase for "Designs for Enabling Collaboration in Human-Machine Teaming via Interactive and Explainable Systems," which is published in NeurIPS 2024 (poster). The presentation video of this work can be found here [FILL].

Authors: Rohan Paleja, Michael Munje, Kimberlee Chang, Reed Jensen, and Matthew Gombolay

Training Agent Models

As mentioned in the paper, we leverage the PantheonRL codebase to train agents and change the agent representation to be an Interpretable Discrete Control Tree instead of the default Neural Network. The IDCT model can be found in ipm/models/idct.py and the domains can be found in overcooked_ai/src/overcooked_ai_py/data/layouts/two_rooms_narrow.layout and overcooked_ai/src/overcooked_ai_py/data/layouts/forced_coordination.layout

Running the interactive policy modification GUI

The main file of this experiment is run_interactive_experiment.py. This file runs the human-subjects study while accounting for which condition the user is in. This file handles all rendering of the Overcooked-AI domain as well as the tree visualization and modification interface.

About

Codebase supporting NeurIPS 2024 Paper: Designs for Enabling Collaboration in Human-Machine Teaming via Interactive and Explainable Systems

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published