BriefGPT.xyz
Sep, 2022
A*学习启发式方法的可微损失函数
A Differentiable Loss Function for Learning Heuristics in A*
HTML
PDF
Leah Chrestien, Tomas Pevny, Antonin Komenda, Stefan Edelkamp
TL;DR
本文讨论用于A*算法启发函数的深度神经网络的优化问题，提出使用L*损失函数可以显著提高迷宫等领域的自动化规划质量和效率。
Abstract
Optimization of
heuristic functions
for the A* algorithm, realized by
deep neural networks
, is usually done by minimizing square root
l* loss
→