arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1206.6398cs.LGstat.ML
收藏

学习参数化技能

Learning Parameterized Skills

University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校);MIT Computer Science and Artificial Intelligence Laboratory(麻省理工学院计算机科学与人工智能实验室)

Bruno Da Silva, George Konidaris, Andrew Barto

摘要:

我们介绍了一种构建技能的方法,能够解决从分布的参数化强化学习问题的任务。该方法从感兴趣的分布中提取示例任务,并使用相应的学习策略来估计技能策略所在的低维分段光滑流形的拓扑。该流形对策略参数如何随着任务参数的变化而变化进行建模。该方法识别组成流形的图表的数量,然后在每个图表中应用非线性回归,通过从任务参数预测策略参数来构建参数化技能。我们评估我们的方法欠驱动的模拟机器人手臂的任务是学习准确地扔飞镖在一个参数化的目标位置。

英文摘要:

We introduce a method for constructing skills capable of solving tasks drawn from a distribution of parameterized reinforcement learning problems. The method draws example tasks from a distribution of interest and uses the corresponding learned policies to estimate the topology of the lower-dimensional piecewise-smooth manifold on which the skill policies lie. This manifold models how policy parameters change as task parameters vary. The method identifies the number of charts that compose the manifold and then applies non-linear regression in each chart to construct a parameterized skill by predicting policy parameters from task parameters. We evaluate our method on an underactuated simulated robotic arm tasked with learning to accurately throw darts at a parameterized target location.

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。

↑