可提示行为：从人类偏好中个性化多目标奖励
Promptable Behaviors: Personalizing Multi-Objective Rewards from Human Preferences
AI总结：
针对具身AI中机器人行为适配多样人类偏好的难题，提出Promptable Behaviors框架，用多目标强化学习训练通用策略，通过三类交互推断偏好，在导航任务中验证了个性化效果。
摘要：
定制机器人行为以匹配多样化人类偏好，是具身AI领域一个尚未得到充分探索的挑战。本文提出Promptable Behaviors（可提示行为）这一新型框架，可助力机器人智能体在复杂环境中高效实现个性化，适配多样人类偏好。我们采用多目标强化学习训练单个策略，使其能适配广泛的偏好范围。我们提出三种不同的人类偏好推断方法，分别利用不同类型的交互：（1）人类演示，（2）轨迹对比的偏好反馈，（3）语言指令。我们在ProcTHOR和RoboTHOR中的个性化物体目标导航与逃离导航任务上评估了所提方法，证明了其可通过提示调整智能体行为，在多种场景下满足人类偏好。项目页面：https://promptable-behaviors.github.io
英文摘要：
Customizing robotic behaviors to be aligned with diverse human preferences is an underexplored challenge in the field of embodied AI. In this paper, we present Promptable Behaviors, a novel framework that facilitates efficient personalization of robotic agents to diverse human preferences in complex environments. We use multi-objective reinforcement learning to train a single policy adaptable to a broad spectrum of preferences. We introduce three distinct methods to infer human preferences by leveraging different types of interactions: (1) human demonstrations, (2) preference feedback on trajectory comparisons, and (3) language instructions. We evaluate the proposed method in personalized object-goal navigation and flee navigation tasks in ProcTHOR and RoboTHOR, demonstrating the ability to prompt agent behaviors to satisfy human preferences in various scenarios. Project page: https://promptable-behaviors.github.io
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。
