Learning differentiable logic programs for abstract visual reasoning

作     者：Shindo, Hikaru Pfanschilling, Viktor Dhami, Devendra Singh Kersting, Kristian 

作者机构：Tech Univ Darmstadt Darmstadt Germany Hessian Ctr hessian AI Darmstadt Germany Tech Univ Darmstadt Ctr Cognit Sci Darmstadt Germany German Ctr Artificial Intelligence DFKI Darmstadt Germany Eindhoven Univ Technol Eindhoven Netherlands 

出 版 物：《MACHINE LEARNING》 (Mach Learn)

年 卷 期：2024年第113卷第11-12期

页      面：8533-8584页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Projekt DEAL AI lighthouse project "SPAICER" [01MK20015E] EU ICT-48 Network of AI Research Excellence Center "TAILOR" (EU Horizon 2020) Collaboration Lab "AI in Construction" (AICO) Department of Mathematics and Computer Science and the Eindhoven Artificial Intelligence Systems Institute 

主　　题：Neuro-symbolic AI Differentiable reasoning Inductive logic programming Graph neural network 

摘      要：Visual reasoning is essential for building intelligent agents that understand the world and perform problem-solving beyond perception. Differentiable forward reasoning has been developed to integrate reasoning with gradient-based machine learning paradigms. However, due to the memory intensity, most existing approaches do not bring the best of the expressivity of first-order logic, excluding a crucial ability to solve abstract visual reasoning, where agents need to perform reasoning by using analogies on abstract concepts in different scenarios. To overcome this problem, we propose NEUro-symbolic Message-pAssiNg reasoNer (NEUMANN), which is a graph-based differentiable forward reasoner, passing messages in a memory-efficient manner and handling structured programs with functors. Moreover, we propose a computationally-efficient structure learning algorithm to perform explanatory program induction on complex visual scenes. To evaluate, in addition to conventional visual reasoning tasks, we propose a new task, visual reasoning behind-the-scenes, where agents need to learn abstract programs and then answer queries by imagining scenes that are not observed. We empirically demonstrate that NEUMANN solves visual reasoning tasks efficiently, outperforming neural, symbolic, and neuro-symbolic baselines.