Code associated with the publication: What model does MuZero learn?
doi:10.4121/a88194f4-45a5-41ae-9cda-0116b78473e5.v1
The doi above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future.
For a link that will always point to the latest version, please use
doi: 10.4121/a88194f4-45a5-41ae-9cda-0116b78473e5
doi: 10.4121/a88194f4-45a5-41ae-9cda-0116b78473e5
Datacite citation style:
He, Jinke; Moerland, Thomas; de Vries, Joery; Oliehoek, Frans (2024): Code associated with the publication: What model does MuZero learn?. Version 1. 4TU.ResearchData. dataset. https://doi.org/10.4121/a88194f4-45a5-41ae-9cda-0116b78473e5.v1
Other citation styles (APA, Harvard, MLA, Vancouver, Chicago, IEEE) available at Datacite
Dataset
licence
MIT
This Dataset contains the code associated with the research paper "What model does MuZero learn?" published at ECAI 2024.
Our research aims to study to what extent models learned by MuZero support policy improvement.
The code here contains scripts to evaluate models learned by MuZero.
Two files implement our policy evaluation and improvement experiments with common functionalities implemented in base.py.
The other scripts are for scaling experiments by automatically generating and launching experiment configurations.
To train MuZero agents, we used https://github.com/YeWR/EfficientZero and https://github.com/werner-duvaud/muzero-general.
history
- 2024-12-11 first online, published, posted
publisher
4TU.ResearchData
format
four of the files are .py files. one is .sh file.
associated peer-reviewed publication
What model does MuZero learn?
organizations
TU Delft, Faculty of Electrical Engineering, Mathematics and Computer Science, Department of Intelligent Systems
DATA
files (6)
- 747 bytesMD5:
3f0e51e1f7caf3eed3afd3bc94fab062
README.rtf - 30,002 bytesMD5:
4bd09d2ed177c9bfe508587956d05e8b
base.py - 36,586 bytesMD5:
99332e739f79a2a83e2ea4fa4fb1559a
experimenter.py - 27,723 bytesMD5:
fa203646dff2b4b4f3d7de859dc2a5e8
launch_exps.sh - 9,490 bytesMD5:
45a195492a2ed69f9ada3286a642ac99
test_policies.py - 26,505 bytesMD5:
506abe085b35d21e1cbcb186bf12df1d
test_value_prediction_error.py -
download all files (zip)
131,053 bytes unzipped