# Outracing champion Gran Turismo drivers with deep reinforcement learning

*Source* · `src-wurman-2022-gt-sophy` · status: **n/a**

| field | value |
|---|---|
| kind | Peer-reviewed paper |
| authors | Peter R. Wurman, et al. |
| venue | Nature 602:223-228 |
| date | 2022-02-09 |
| doi | 10.1038/s41586-021-04357-7 |
| accessed | 2026-10-24 |
| notes | Sony AI; model-free deep RL with mixed-scenario training and a reward shaped for sportsmanship rules; beat four top Gran Turismo drivers head to head. |

## Backlinks (computed)

**cited by:** [Deep reinforcement learning policies for NPC opponents and navigation](../technique/deep-rl-npc-policies.md), [Gran Turismo 7](../game/gran-turismo-7.md)

## Citation

Peter R. Wurman, et al. (2022-02-09). *Outracing champion Gran Turismo drivers with deep reinforcement learning*. Nature 602:223-228 https://doi.org/10.1038/s41586-021-04357-7

---
Provenance: created in run `2026-10-24-techniques-rl-design`.