moporgic/TDL2048+

TDL2048+ is a highly optimized temporal difference (TD) learning framework for 2048. This repository provides publicly available baseline models for use with TDL2048+. For instructions on using these models, please refer to the main program repository: moporgic/TDL2048+.

Models

The models in this repository are n-tuple networks, a network architecture first applied to 2048 by Szubert and Jaśkowski. They were trained using a variant of temporal difference learning, optimistic temporal difference learning. This repository includes five commonly used network structures:

4x6patt.w: A 4×6-tuple network proposed by Yeh et al.
5x6patt.w: A 5×6-tuple network proposed by Jaśkowski.
6x6patt.w: A 6×6-tuple network proposed by Matsuzaki.
7x6patt.w: A 7×6-tuple network proposed by Matsuzaki.
8x6patt.w: A 8×6-tuple network proposed by Matsuzaki.

License

TDL2048+ and the released models are licensed under the MIT License.

Show the license

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

References

M. Szubert and W. Jaśkowski, “Temporal difference learning of N tuple networks for the game 2048,” in Proc. 2014 IEEE Conf. Comput. Intell. Games, Dortmund, Germany, 2014, pp. 1–8, doi: 10.1109/CIG.2014.6932907.
H. Guei, L.-P. Chen and I-C. Wu, “Optimistic Temporal Difference Learning for 2048,” in IEEE Trans. Games, vol. 14, no. 3, pp. 478–487, Sep. 2022, doi: 10.1109/TG.2021.3109887. [Online]. Available: arXiv:2111.11090.
K.-H. Yeh, I-C. Wu, C.-H. Hsueh, C.-C. Chang, C.-C. Liang, and H. Chiang, “Multistage temporal difference learning for 2048-like games,” IEEE Trans. Comput. Intell. AI Games, vol. 9, no. 4, pp. 369–380, Dec. 2017, doi: 10.1109/TCIAIG.2016.2593710.
W. Jaśkowski, “Mastering 2048 with delayed temporal coherence learning, multistage weight promotion, redundant encoding and carousel shaping,” IEEE Trans. Games, vol. 10, no. 1, pp. 3–14, Mar. 2018, doi: 10.1109/TCIAIG.2017.2651887.
K. Matsuzaki, “Systematic selection of N tuple networks with consideration of interinfluence for game 2048,” in Proc. 21st Int. Conf. Technol. Appl. Artif. Intell., Hsinchu, Taiwan, 2016, pp. 186–193, doi: 10.1109/TAAI.2016.7880154.