arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2001.04465cs.ROcs.AIcs.HCcs.LGstat.ML
收藏

LESS是更多:重新思考人类行为的概率模型

LESS is More: Rethinking Probabilistic Models of Human Behavior

University of California, Berkeley(加州大学伯克利分校)

Andreea Bobu, Dexter R. R. Scobee, Jaime F. Fisac, S. Shankar Sastry, Anca D. Dragan

摘要:

机器人需要人类行为模型来推断人类的目标和偏好,并预测人们会做什么。一个常见的模型是玻尔兹曼噪声-理性决策模型,它假设人们近似地优化奖励函数,并选择与他们的指数奖励成比例的轨迹。虽然这个模型在各种机器人领域都取得了成功,但它的根源在于计量经济学,以及对不同离散选项之间的决策建模,每个选项都有自己的效用或奖励。相比之下,人类轨迹位于连续空间中,具有影响奖励函数的连续值特征。我们建议,现在是时候重新思考玻尔兹曼模型,并从头开始设计它,以在这样的轨迹空间上运行。我们引入了一个模型,明确说明轨迹之间的距离,而不仅仅是他们的奖励。不是每个轨迹独立地影响决策,而是类似的轨迹一起影响决策。我们首先展示了我们的模型在用户研究中更好地解释了人类行为。然后,我们分析了这对机器人推理的影响,首先是在玩具环境中,我们有地面真理,并找到更准确的推理,最后是从用户演示中学习的7自由度机器人手臂。

英文摘要:

Robots need models of human behavior for both inferring human goals and preferences, and predicting what people will do. A common model is the Boltzmann noisily-rational decision model, which assumes people approximately optimize a reward function and choose trajectories in proportion to their exponentiated reward. While this model has been successful in a variety of robotics domains, its roots lie in econometrics, and in modeling decisions among different discrete options, each with its own utility or reward. In contrast, human trajectories lie in a continuous space, with continuous-valued features that influence the reward function. We propose that it is time to rethink the Boltzmann model, and design it from the ground up to operate over such trajectory spaces. We introduce a model that explicitly accounts for distances between trajectories, rather than only their rewards. Rather than each trajectory affecting the decision independently, similar trajectories now affect the decision together. We start by showing that our model better explains human behavior in a user study. We then analyze the implications this has for robot inference, first in toy environments where we have ground truth and find more accurate inference, and finally for a 7DOF robot arm learning from user demonstrations.

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。

↑