• Shenghui Chen, Daniel Fried, Ufuk Topcu
• International Joint Conference on Artificial Intelligence (IJCAI), Human-Centred Artificial Intelligence track, 2024

If you use this work, please cite:

@inproceedings{chen2024human,
    title     = {Human-Agent Cooperation in Games under Incomplete Information through Natural Language Communication},
    author    = {Chen, Shenghui and Fried, Daniel and Topcu, Ufuk},
    booktitle = {Proceedings of the Thirty-Third International Joint Conference on
               Artificial Intelligence ({IJCAI-24}), Human-Centred AI},
    pages     = {7833--7841},
    year      = {2024},
    doi       = {10.24963/ijcai.2024/867}
}

🌟 Developing autonomous agents to strategize and cooperate with humans using natural language is challenging. Our testbed, Gnomes at Night 🎮, is a maze game where two players alternately control a token to achieve a common goal with incomplete information.

✍️ We introduce a shared-control game, where two players collectively control a token ♟️ in alternating turns to achieve a common objective under incomplete information. We formulate a policy synthesis problem for an autonomous agent in this game with a human as the other player.

🤖 To solve this problem, we propose a communication-based approach comprising a language module and a planning module. The language module translates natural language messages into and from a finite set of flags, a compact representation defined to capture player intents. The planning module leverages these flags to compute a policy using an asymmetric information-set Monte Carlo tree search with flag exchange (AISMCTS-F) algorithm we present.

                      

It is recommended to create a Python virtual environment first:

python -m venv venv
source venv/bin/activate

Then install the dependencies:

pip install -e .

To use the OpenAI API, you need to add your API key 🔑 in a .env file located in the root directory. Create the .env file with the following content:

OPENAI_API_KEY=[your openai api key]

This API key will be used in the src/shared_control_language/aismcts_language_utils.py file. Make sure to replace [your openai api key] with your actual OpenAI API key.

To test this algorithm in a 9x9 Gnomes at Night environment called GnomesAtNightEnv9A by interacting with an agent in the terminal, run the following command:

python scripts/test_aismcts_language.py --round 1 --explore 100 --render

The command line arguments include

• --round: Specifies the round number (in [1, 2, 3, 4, 5], default is 1).
• --explore: Sets the number of iterations (exploration constant) in MCTS (must be an integer, default is 100).
• --render: Enables rendering of the environment (use this flag only when rendering is desired).

At the end of gameplay, you should see logging output 📝 similar to the following:

===============================================
GnomesAtNightEnv9A Play Summary
-----------------------------------------------
Round         1
MCTS Explore  100
Render        False
-----------------------------------------------
Total Reward  17.0
Total Steps   6
Side Info     {}
===============================================