arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1811.06151cs.AIcs.RO
收藏

正交策略梯度及自动驾驶应用

Orthogonal Policy Gradient and Autonomous Driving Application

China Institute of Atomic Energy(中国原子能科学研究院)

Mincong Luo, Yin Tong, Jiachi Liu

摘要:

深度强化学习的一个较少解决的问题是缺乏基于新状态和新目标的泛化能力,对于复杂任务,需要给出正确的策略并评估当前状态的所有可能动作。幸运的是,深度强化学习在两个子问题上都取得了巨大的进步:给出正确的策略,并根据状态评估所有行动。本文提出了一种称为正交策略梯度下降(OPGD)的方法,使Agent能够根据当前状态和动作集学习策略梯度,从而学习具有泛化能力的策略网络。在三维自动驾驶环境TORCS上,将该方法与基线模型进行了比较,并给出了详细的实验结果分析和证明。

英文摘要:

One less addressed issue of deep reinforcement learning is the lack of generalization capability based on new state and new target, for complex tasks, it is necessary to give the correct strategy and evaluate all possible actions for current state. Fortunately, deep reinforcement learning has enabled enormous progress in both subproblems: giving the correct strategy and evaluating all actions based on the state. In this paper we present an approach called orthogonal policy gradient descent(OPGD) that can make agent learn the policy gradient based on the current state and the actions set, by which the agent can learn a policy network with generalization capability. we evaluate the proposed method on the 3D autonomous driving enviroment TORCS compared with the baseline model, detailed analyses of experimental results and proofs are also given.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑