Verification Learning: Make Unsupervised Neuro-Symbolic System Feasible

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Jia, Lin-Han, Hu, Wen-Chao, Shao, Jie-Jing, Guo, Lan-Zhe, Li, Yu-Feng
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909651201687552
author Jia, Lin-Han
Hu, Wen-Chao
Shao, Jie-Jing
Guo, Lan-Zhe
Li, Yu-Feng
author_facet Jia, Lin-Han
Hu, Wen-Chao
Shao, Jie-Jing
Guo, Lan-Zhe
Li, Yu-Feng
contents The current Neuro-Symbolic (NeSy) Learning paradigm suffers from an over-reliance on labeled data, so if we completely disregard labels, it leads to less symbol information, a larger solution space, and more shortcuts-issues that current Nesy systems cannot resolve. This paper introduces a novel learning paradigm, Verification Learning (VL), which addresses this challenge by transforming the label-based reasoning process in Nesy into a label-free verification process. VL achieves excellent learning results solely by relying on unlabeled data and a function that verifies whether the current predictions conform to the rules. We formalize this problem as a Constraint Optimization Problem (COP) and propose a Dynamic Combinatorial Sorting (DCS) algorithm that accelerates the solution by reducing verification attempts, effectively lowering computational costs and introduce a prior alignment method to address potential shortcuts. Our theoretical analysis points out which tasks in Nesy systems can be completed without labels and explains why rules can replace infinite labels for some tasks, while for others the rules have no effect. We validate the proposed framework through several fully unsupervised tasks including addition, sort, match, and chess, each showing significant performance and efficiency improvements.
format Preprint
id arxiv_https___arxiv_org_abs_2503_12917
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Verification Learning: Make Unsupervised Neuro-Symbolic System Feasible
Jia, Lin-Han
Hu, Wen-Chao
Shao, Jie-Jing
Guo, Lan-Zhe
Li, Yu-Feng
Artificial Intelligence
The current Neuro-Symbolic (NeSy) Learning paradigm suffers from an over-reliance on labeled data, so if we completely disregard labels, it leads to less symbol information, a larger solution space, and more shortcuts-issues that current Nesy systems cannot resolve. This paper introduces a novel learning paradigm, Verification Learning (VL), which addresses this challenge by transforming the label-based reasoning process in Nesy into a label-free verification process. VL achieves excellent learning results solely by relying on unlabeled data and a function that verifies whether the current predictions conform to the rules. We formalize this problem as a Constraint Optimization Problem (COP) and propose a Dynamic Combinatorial Sorting (DCS) algorithm that accelerates the solution by reducing verification attempts, effectively lowering computational costs and introduce a prior alignment method to address potential shortcuts. Our theoretical analysis points out which tasks in Nesy systems can be completed without labels and explains why rules can replace infinite labels for some tasks, while for others the rules have no effect. We validate the proposed framework through several fully unsupervised tasks including addition, sort, match, and chess, each showing significant performance and efficiency improvements.
title Verification Learning: Make Unsupervised Neuro-Symbolic System Feasible
topic Artificial Intelligence
url https://arxiv.org/abs/2503.12917