arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.08502cs.ROcs.SYeess.SY

通过加权时间逻辑实现安全且最优的偏好学习:应用于机器人与F1的应用

Safe and Optimal Learning from Preferences via Weighted Temporal Logic with Applications in Robotics and Formula 1

Ruya Karagulle, Cristian-Ioan Vasile, Necmiye Ozay

AI总结:

本文提出基于加权时间逻辑的安全最优偏好学习方法,应用于机器人与F1领域,通过结构修剪和对数转换提升安全性与效率。

摘要:

自主系统日益依赖人类反馈来对齐其行为,这些反馈以成对比较、排名或示范的形式表达。尽管现有方法可以适应行为,但在安全关键领域往往无法保证安全性。我们提出了一种安全保证、最优且高效的解决从偏好、排名或示范中学习问题的方法,使用加权信号时间逻辑(WSTL)。当直接实施WSTL学习问题时,会导致需要学习的权重出现多线性约束。通过引入结构修剪和对数转换程序,我们减少了问题规模,并将其重新表述为混合整数线性规划问题,同时保持安全性保证。在机器人导航和现实世界F1数据上的实验表明,该方法能够捕捉细微的偏好并建模复杂的任务目标。

英文摘要:

Autonomous systems increasingly rely on human feedback to align their behavior, expressed as pairwise comparisons, rankings, or demonstrations. While existing methods can adapt behaviors, they often fail to guarantee safety in safety-critical domains. We propose a safety-guaranteed, optimal, and efficient approach for solving the learning problem from preferences, rankings, or demonstrations using Weighted Signal Temporal Logic (WSTL). WSTL learning problems, when implemented naively, lead to multi-linear constraints in the weights to be learned. By introducing structural pruning and log-transform procedures, we reduce the problem size and recast it as a Mixed-Integer Linear Program while preserving safety guarantees. Experiments on robotic navigation and real-world Formula 1 data demonstrate that the method captures nuanced preferences and models complex task objectives.

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。

↑