> ML_LITERATURE // WURMAN-2022-OUTRACING-CHAMPION-GRAN-TURISMO-DRIVERS-GT-SOPHY_v1.0
Outracing champion Gran Turismo drivers with deep reinforcement learning (GT Sophy)
Peter R. Wurman, Samuel Barrett, Kenta Kawamoto, James MacGlashan, Kaushik Subramanian, Thomas J. Walsh, Roberto Capobianco, Alisa Devlic, Franziska Eckert, Florian Fuchs, Gandhali Joshi, Hadrien Kras, Bo Lu, Matthew Overman, Khai Vu, Hidenori Ysumoto, Shunsuke Aoki, Raymond Mooney, Peter Stone, Hiroaki Kitano · Nature (2022)
seminal-architecture2022foundationalthirdPartyReproduced
Principal Contribution
Trained autonomous agents combining quantile regression soft actor-critic with mixed scenario training, defeating the best human esports drivers in Gran Turismo Sport.
Operational Relevance
Serves as qualified theoretical and systems foundation for task-game-playing, task-continuous-control.
Assumptions
- Markovian state dynamics and stationary reward functions hold in target evaluation environments
Limitations
- Sample efficiency, exploration stability, and real-world sim-to-real transfer gaps require specialized tuning
Connected Algorithms, Architectures & Tools
Related Algorithms:
Related Architectures:
Implementing Libraries: