arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1709.07979cs.ROcs.AIcs.LG
收藏

基于梯度引导策略专业化的多任务学习

Multi-task Learning with Gradient Guided Policy Specialization

Georgia Institute of Technology(佐治亚理工学院)

Wenhao Yu, C. Karen Liu, Greg Turk

摘要:

我们提出了一种方法,有效地学习多个相关的机器人运动技能的控制策略。我们的方法包括两个阶段,联合培训和专业化培训。在联合训练阶段,用最少的信息训练神经网络策略以消除运动技能的歧义。这迫使策略学习不同任务的公共表示。然后,在专业化训练阶段,我们根据衡量多个任务之间不一致性的每个权重度量有选择地划分策略的权重。通过分割控制策略的一部分,可以进一步训练它来专门处理每个任务。为了在学习过程中更新控制策略,我们使用广义优势函数信赖域策略优化(TRPOGAE)。我们提出了一个修改的梯度更新阶段的TRPO,以更好地适应多任务的学习场景。我们评估了我们在模拟中的三个连续运动技能学习问题上的方法:1)运动任务,其中训练具有相当大的形状和尺寸差异的三个单腿足式机器人向前跳跃,2)操纵任务,其中训练具有不同尺寸和关节类型的三个机器人操纵器到达3D空间中的不同位置,以及3)两条腿的足式机器人的运动,其一条腿的运动范围以不同的方式受到限制。我们将我们的训练方法与三个基线进行比较。第一个基线只对策略进行联合训练,第二个基线为每个任务训练独立的策略,最后一个基线随机选择要分割的权重。我们表明,我们的方法比每种基线方法都更有效地学习。

英文摘要:

We present a method for efficient learning of control policies for multiple related robotic motor skills. Our approach consists of two stages, joint training and specialization training. During the joint training stage, a neural network policy is trained with minimal information to disambiguate the motor skills. This forces the policy to learn a common representation of the different tasks. Then, during the specialization training stage we selectively split the weights of the policy based on a per-weight metric that measures the disagreement among the multiple tasks. By splitting part of the control policy, it can be further trained to specialize to each task. To update the control policy during learning, we use Trust Region Policy Optimization with Generalized Advantage Function (TRPOGAE). We propose a modification to the gradient update stage of TRPO to better accommodate multi-task learning scenarios. We evaluate our approach on three continuous motor skill learning problems in simulation: 1) a locomotion task where three single legged robots with considerable difference in shape and size are trained to hop forward, 2) a manipulation task where three robot manipulators with different sizes and joint types are trained to reach different locations in 3D space, and 3) locomotion of a two-legged robot, whose range of motion of one leg is constrained in different ways. We compare our training method to three baselines. The first baseline uses only joint training for the policy, the second trains independent policies for each task, and the last randomly selects weights to split. We show that our approach learns more efficiently than each of the baseline methods.

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。

↑