BriefGPT.xyz
Jun, 2024
从规约中归纳推理的强化学习
Inductive Generalization in Reinforcement Learning from Specifications
HTML
PDF
Vignesh Subramanian, Rohit Kushwah, Subhajit Roy, Suguman Bansal
TL;DR
我们提出了一个新颖的归纳一般化框架，用于从逻辑规范中进行强化学习。该框架通过利用归纳任务之间的关系，学习生成适应归纳任务实例的策略生成器，以实现对长期任务中未见策略的广义化。
Abstract
We present a novel
inductive generalization framework
for
rl
from
logical specifications
. Many interesting tasks in
→