arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2307.04895cs.AIcs.LG
收藏

使用 Recurrent Transformer 学习求解约束满足问题

Learning to Solve Constraint Satisfaction Problems with Recurrent Transformer

Arizona State University(亚利桑那州立大学);Samsung Research(三星研究院)

Zhun Yang, Adam Ishay, Joohyung Lee

AI总结:

本文提出 Recurrent Transformer,以端到端方式学习求解 CSPs,并支持视觉约束推理、符号接地、样本高效与半监督学习。

摘要:

约束满足问题(CSPs)旨在寻找满足给定约束的变量取值。本文表明,扩展了循环机制的 Transformer 是一种可行的端到端学习求解 CSPs 的方法,并且相较 Graph Neural Networks、SATNet 以及若干神经符号模型等最先进方法具有明显优势。借助 Transformer 处理视觉输入的能力,所提出的 Recurrent Transformer 可以直接应用于视觉约束推理问题,同时成功解决符号接地问题。我们还展示了如何在 Transformer 的归纳学习中利用离散约束的演绎知识,从而为 CSPs 实现样本高效学习和半监督学习。

英文摘要:

Constraint satisfaction problems (CSPs) are about finding values of variables that satisfy the given constraints. We show that Transformer extended with recurrence is a viable approach to learning to solve CSPs in an end-to-end manner, having clear advantages over state-of-the-art methods such as Graph Neural Networks, SATNet, and some neuro-symbolic models. With the ability of Transformer to handle visual input, the proposed Recurrent Transformer can straightforwardly be applied to visual constraint reasoning problems while successfully addressing the symbol grounding problem. We also show how to leverage deductive knowledge of discrete constraints in the Transformer's inductive learning to achieve sample-efficient learning and semi-supervised learning for CSPs.

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。

↑