Column Networks for Collective Classification

TL;DR

This paper introduces Column Networks (CLN), a deep learning model designed for collective classification in relational data, achieving higher accuracy efficiently across various real-world applications.

Contribution

It presents a novel deep learning architecture that encodes multi-relational data, supports complex dependencies, and is computationally efficient for collective classification tasks.

Findings

CLN outperforms state-of-the-art methods in multiple applications.

CLN efficiently models long-range and higher-order dependencies.

CLN demonstrates superior accuracy in real-world collective classification tasks.

Abstract

Relational learning deals with data that are characterized by relational structures. An important task is collective classification, which is to jointly classify networked objects. While it holds a great promise to produce a better accuracy than non-collective classifiers, collective classification is computational challenging and has not leveraged on the recent breakthroughs of deep learning. We present Column Network (CLN), a novel deep learning model for collective classification in multi-relational domains. CLN has many desirable theoretical properties: (i) it encodes multi-relations between any two instances; (ii) it is deep and compact, allowing complex functions to be approximated at the network level with a small set of free parameters; (iii) local and relational features are learned simultaneously; (iv) long-range, higher-order dependencies between instances are supported naturally; and (v) crucially, learning and inference are efficient, linear in the size of the network and the number of relations. We evaluate CLN on multiple real-world applications: (a) delay prediction in software projects, (b) PubMed Diabetes publication classification and (c) film genre classification. In all applications, CLN demonstrates a higher accuracy than state-of-the-art rivals.