arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1706.07138cs.LG
收藏

使用深度层次网络生成长期轨迹

Generating Long-term Trajectories Using Deep Hierarchical Networks

Caltech(加州理工学院);STATS(STATS公司)

Stephan Zheng, Yisong Yue, Patrick Lucey

摘要:

我们研究的问题,建模时空轨迹在很长的时间范围内使用专家演示。例如,在体育运动中,智能体经常选择具有长期目标的动作序列,例如实现某个战略位置。传统的策略学习方法,例如基于马尔可夫决策过程的方法,通常无法在这种高维状态空间中学习内聚的长期行为,并且仅在近视建模导致期望的行为时才有效。关键的困难在于,传统的方法是“浅”的模型,只学习一个单一的国家行动的政策。相反,我们提出了一个分层的政策类,自动的原因,长期和短期的目标,我们作为一个分层的神经网络实例化。我们展示了我们的方法在学习模仿演示篮球轨迹的案例研究,并表明,它产生显着更现实的轨迹相比,由专业体育分析师判断的非层次基线。

英文摘要:

We study the problem of modeling spatiotemporal trajectories over long time horizons using expert demonstrations. For instance, in sports, agents often choose action sequences with long-term goals in mind, such as achieving a certain strategic position. Conventional policy learning approaches, such as those based on Markov decision processes, generally fail at learning cohesive long-term behavior in such high-dimensional state spaces, and are only effective when myopic modeling lead to the desired behavior. The key difficulty is that conventional approaches are "shallow" models that only learn a single state-action policy. We instead propose a hierarchical policy class that automatically reasons about both long-term and short-term goals, which we instantiate as a hierarchical neural network. We showcase our approach in a case study on learning to imitate demonstrated basketball trajectories, and show that it generates significantly more realistic trajectories compared to non-hierarchical baselines as judged by professional sports analysts.

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。

↑