Abstract
We introduce a framework for model learning and planning in stochastic domains with continuous state and action spaces and non-Gaussian transition models. It is efficient because (1) local models are estimated only when the planner requires them; (2) the planner focuses on the most relevant states to the current planning problem; and (3) the planner focuses on the most informative and/or high-value actions. Our theoretical analysis shows the validity and asymptotic optimality of the proposed approach. Empirically, we demonstrate the effectiveness of our algorithm on a simulated multi-modal pushing problem.
Abstract (translated by Google)
我们在具有连续状态和动作空间以及非高斯过渡模型的随机域中引入了模型学习和规划的框架。它是有效的，因为（1）只有当规划者需要时才估计本地模型; （2）规划者关注当前规划问题中最相关的状态; （3）计划者关注最具信息性和/或高价值的行动。我们的理论分析表明了该方法的有效性和渐近最优性。根据经验，我们证明了我们的算法在模拟多模态推力问题上的有效性。
URL
https://arxiv.org/abs/1607.07762