行为克隆中面向分布外泛化的问题空间变换
Problem Space Transformations for Out-of-Distribution Generalisation in Behavioural Cloning
AI总结:
本文提出通过问题空间变换提升行为克隆在分布外泛化中的性能,通过实验验证了姿态等价性和局部性对动作预测的改进作用。
摘要:
行为克隆与神经网络的结合推动了机器人操作的显著进展。由于这些算法可能需要为每个感兴趣的任务提供大量的示例,因此在复杂场景中,它们仍然本质上是低效的,有限的数据集几乎无法覆盖状态空间。因此,剩下的挑战之一是分布外(OOD)泛化,即预测低概率状态下的正确动作的能力。当控制系统被视为黑箱,忽略其物理特性时,这一问题会更加严重。本文强调了机器人操作中的普遍特性,特别是姿态等价性和局部性。我们研究了问题空间选择对行为克隆策略OOD性能的影响,以及如何利用操作的特征属性所产生的变换来改进其性能。通过受控、模拟和现实世界实验,我们实证表明,这些变换使使用标准MLP基础的一步动作预测或扩散基础的动作序列预测的行为克隆策略能够更好地泛化到某些OOD问题实例。代码可在https://github.com/kirandoshi/pst_ood_gen上获取。
英文摘要:
The combination of behavioural cloning and neural networks has driven significant progress in robotic manipulation. As these algorithms may require a large number of demonstrations for each task of interest, they remain fundamentally inefficient in complex scenarios, in which finite datasets can hardly cover the state space. One of the remaining challenges is thus out-of-distribution (OOD) generalisation, i.e. the ability to predict correct actions for states with a low likelihood with respect to the state occupancy induced by the dataset. This issue is aggravated when the system to control is treated as a black-box, ignoring its physical properties. This work highlights widespread properties of robotic manipulation, specifically pose equivariance and locality. We investigate the effect of the choice of problem space on OOD performance of BC policies and how transformations arising from characteristic properties of manipulation can be employed for its improvement. Through controlled, simulated and real-world experiments, we empirically demonstrate that these transformations allow behaviour cloning policies, using either standard MLP-based one-step action prediction or diffusion-based action-sequence prediction, to generalise better to certain OOD problem instances. Code is available at https://github.com/kirandoshi/pst_ood_gen.
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。