Abstract: Reinforcement learning (RL) is a promising approach for end-to-end autonomous driving, but its practical deployment remains challenging due to low sample efficiency and sensitivity to reward ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results