Why is RL hard for autonomous driving?
hardAnswer
- (1) Safety: cannot explore in real traffic.
- (2) Long-tail rare events matter (fatal edge cases).
- (3) Multi-agent interactions (other drivers, pedestrians) with non-stationary policies.
- (4) Partial observability.
- (5) Reward specification difficulty.
- Most industry uses SL / imitation learning + planning stack (Waymo, Cruise).
- RL used mostly in simulation for policy fine-tuning, closed-loop training.
- Recent: Tesla's neural planners are hybrid IL + RL.
Check yourself — multiple choice
- Same as games
- Safety (no exploration) + long-tail edge cases + multi-agent non-stationary + partial obs + reward specification hard; industry uses SL/IL + planning; RL in sim + Tesla hybrid IL+RL
- Random
- Not real
AD hard: safety + long-tail + multi-agent + partial obs; industry SL/IL + sim RL.
#applications#safety
Practise Reinforcement Learning
214 interview questions in this topic.