Most estimators implemented for the video game industry cannot obtain reliable initial estimates nor guarantee comparability between distant estimates. TrueSkill Through Time solves all these problems by modeling the entire history of activities using a single Bayesian network. This algorithm requires a few iterations to converge, allowing millions of observations to be analyzed using any low-end computer.
To support the use of reliable learning estimators, we provide the first implementations of TrueSkill Through Time for Julia, Python and R.
A complete scientific documentation allows scientists to make sense of all epistemological and technical aspects of the estimation process.
Scientific article
You can find the full scientific documentation of TrueSkill Through Time packages at:
- English. Last version 2021-07-26
- Español. Última versión 2021-07-26
Packages
- Julia Package: https://github.com/glandfried/TrueSkillThroughTime.jl
- Python Package: https://github.com/glandfried/TrueSkillThroughTime.py
- R package: https://github.com/glandfried/TrueSkillThroughTime.R
Computational details
Our Python package solves individual events ten times faster than the original trueskill 0.4.5 (Lee 2012) package. In turn, our Julia package converge a history of events ten times faster than our Python package. In contrast, our R package is slower than the other packages, including the original trueskill 0.4.5 package.
Acknowledgments
Special thanks to Heungsub Lee for having published the basic TrueSkill model in Python.
