Saved in:
Bibliographic Details
Main Authors: Fu, Jingwen, Zhang, Zhizheng, Lu, Yan, Zheng, Nanning
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.11743
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929348749033472
author Fu, Jingwen
Zhang, Zhizheng
Lu, Yan
Zheng, Nanning
author_facet Fu, Jingwen
Zhang, Zhizheng
Lu, Yan
Zheng, Nanning
contents Compositional Generalization (CG) embodies the ability to comprehend novel combinations of familiar concepts, representing a significant cognitive leap in human intellectual advancement. Despite its critical importance, the deep neural network (DNN) faces challenges in addressing the compositional generalization problem, prompting considerable research interest. However, existing theories often rely on task-specific assumptions, constraining the comprehensive understanding of CG. This study aims to explore compositional generalization from a task-agnostic perspective, offering a complementary viewpoint to task-specific analyses. The primary challenge is to define CG without overly restricting its scope, a feat achieved by identifying its fundamental characteristics and basing the definition on them. Using this definition, we seek to answer the question "what does the ultimate solution to CG look like?" through the following theoretical findings: 1) the first No Free Lunch theorem in CG, indicating the absence of general solutions; 2) a novel generalization bound applicable to any CG problem, specifying the conditions for an effective CG solution; and 3) the introduction of the generative effect to enhance understanding of CG problems and their solutions. This paper's significance lies in providing a general theory for CG problems, which, when combined with prior theorems under task-specific scenarios, can lead to a comprehensive understanding of CG.
format Preprint
id arxiv_https___arxiv_org_abs_2405_11743
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A General Theory for Compositional Generalization
Fu, Jingwen
Zhang, Zhizheng
Lu, Yan
Zheng, Nanning
Machine Learning
Compositional Generalization (CG) embodies the ability to comprehend novel combinations of familiar concepts, representing a significant cognitive leap in human intellectual advancement. Despite its critical importance, the deep neural network (DNN) faces challenges in addressing the compositional generalization problem, prompting considerable research interest. However, existing theories often rely on task-specific assumptions, constraining the comprehensive understanding of CG. This study aims to explore compositional generalization from a task-agnostic perspective, offering a complementary viewpoint to task-specific analyses. The primary challenge is to define CG without overly restricting its scope, a feat achieved by identifying its fundamental characteristics and basing the definition on them. Using this definition, we seek to answer the question "what does the ultimate solution to CG look like?" through the following theoretical findings: 1) the first No Free Lunch theorem in CG, indicating the absence of general solutions; 2) a novel generalization bound applicable to any CG problem, specifying the conditions for an effective CG solution; and 3) the introduction of the generative effect to enhance understanding of CG problems and their solutions. This paper's significance lies in providing a general theory for CG problems, which, when combined with prior theorems under task-specific scenarios, can lead to a comprehensive understanding of CG.
title A General Theory for Compositional Generalization
topic Machine Learning
url https://arxiv.org/abs/2405.11743