Learned heuristics have recently become a competitive alternative to traditional domain-independent heuristics for satisficing planning. Existing approaches, however, focus on improving search guidance rather than guaranteeing admissibility, which makes them unsuitable for optimal classical planning. We present the first method for learning domain-dependent heuristics that are admissible by design and thus preserve the optimality guarantees of A* search. Instead of learning a direct mapping from states to heuristic values, we learn to construct abstractions that induce admissible heuristics. We use an LLM-driven evolutionary program-synthesis framework to obtain, for each domain, a program that produces a pattern collection for any task in that domain, and we combine the resulting patterns admissibly via saturated cost partitioning. Empirically, the learned programs encode interpretable domain-specific insights, run with negligible overhead at test time and yield heuristics that match the coverage of state-of-the-art domain-independent baselines on several domains while evaluating each state substantially faster.
German Research Center for Artificial Intelligence (DFKI)(德国人工智能研究中心);Saarland University(萨尔兰大学);Center for European Research in Trusted Artificial Intelligence (CERTAIN)(欧洲可信人工智能研究中心);Ben-Gurion University of the Negev(内盖夫本-古里安大学)
NutriGen: Personalized Meal Plan Generator Leveraging Large Language Models to Enhance Dietary and Nutritional Adherence
College of Health Solutions, Arizona State University(亚利桑那州立大学健康解决方案学院);School of Computing and Augmented Intelligence, Arizona State University(亚利桑那州立大学计算与增强智能学院)
CLOVER: Closed-Loop Value Estimation and Ranking for End-to-End Autonomous Driving Planning
Department of Automation, University of Science and Technology of China(中国科学技术大学自动化系);Institute for AI Industry Research, Tsinghua University(清华大学人工智能产业研究院);School of Electronic Information Engineering, Beihang University(北航电子信息技术学院);National College for Excellent Engineers, Beihang University(北航卓越工程师学院)
MultiZenoTravel: a Tunable Benchmark for Multi-Objective Planning with Known Pareto Front
YData Lab Inc.(YData实验室公司);Poznan University of Technology(波兹南理工大学);Inria Saclay(法国国家信息与自动化研究所萨克莱分部);Universite Paris-Saclay(巴黎萨克雷大学);Institut Pasteur(巴斯德研究所);Université Paris Cité(巴黎西岱大学)
Fast Downward是一个经典的基于启发式搜索的规划系统。它可以处理PDDL2.2的命题片段中编码的一般确定性规划问题,包括ADL条件和效果以及派生谓词(公理)等高级功能。与HSP和FF等其他著名的规划器一样,Fast Downward是一种渐进式规划器,在向前的方向上搜索规划任务的世界状态空间。然而,与其他PDDL规划系统不同,Fast Downward不直接使用规划任务的命题PDDL表示。相反,输入首先被翻译成一种替代的表示称为多值规划任务,这使得许多命题规划任务的隐式约束显式。利用这种替代表示,Fast Downward使用规划任务的分层分解来计算其启发式函数,称为因果图启发式,这与传统的HSP类启发式算法非常不同,该算法基于忽略操作员的负面相互作用。本文详细介绍了求解多值规划问题的FastDownwards方法。我们将先前对因果图启发式的讨论扩展到涉及公理和条件效应的任务,并提出了一些用于快速向下最佳优先搜索算法的搜索控制的新技术:优选的操作符将有帮助的动作的思想从局部搜索转移到全局最佳优先搜索,启发式函数的延迟评估减轻了大的分支因子对搜索性能的负面影响,而多启发式最佳优先搜索以正交方式将多个启发式评估函数组合在单个搜索算法内。我们还描述了有效的数据结构快速状态扩展(后继生成器和公理评估器),并提出了一种新的非启发式搜索算法称为集中迭代拓宽搜索,它利用编码在因果图中的…
Efficient Heuristic Generation for Robot Path Planning with Recurrent Generative Model
Southern University of Science and Technology(南方科技大学);The Chinese University of Hong Kong(香港中文大学);Shenzhen Research Institute of the Chinese University of Hong Kong(香港中文大学深圳研究院)