Learning-Based Approaches for Graph Problems: A Survey

Abstract

Over the years, many graph problems specifically those in NP-complete are studied by a wide range of researchers. Some famous examples include graph colouring, travelling salesman problem and subgraph isomorphism. Most of these problems are typically addressed by exact algorithms, approximate algorithms and heuristics. There are however some drawback for each of these methods. Recent studies have employed learning-based frameworks such as machine learning techniques in solving these problems, given that they are useful in discovering new patterns in structured data that can be represented using graphs. This research direction has successfully attracted a considerable amount of attention. In this survey, we provide a systematic review mainly on classic graph problems in which learning-based approaches have been proposed in addressing the problems. We discuss the overview of each framework, and provide analyses based on the design and performance of the framework. Some potential research questions are also suggested. Ultimately, this survey gives a clearer insight and can be used as a stepping stone to the research community in studying problems in this field.

Graph neural network inspired algorithm for unsupervised network community detection

“…Overall, while no single heuristic is the best solution for all the cases, a GNNS algorithm often finds a plausible solution, sometimes the best-known one, and provides a flexible parameter-controlled trade-off between speed and performance ranging from the fastest to the close-to-optimal performance, which makes it a valuable addition to an existing collection of algorithms. More importantly, since this simple GNN-style heuristic can perform comparably to the state-of-the-art, that serves as a proof-of-concept for considering more sophisticated GNN architectures, configurations, and learning techniques that could provide further improvement in solving community detection problem as well as other complex network optimization problems like minimum vertex cover, maximal independent set Li et al (2018), clique partitioning, correlation clustering Jung and Keuper (2022), spectral clustering Bianchi et al (2020), and others Bengio et al (2021), Yow and Luo (2022).…”

An Analysis on Research Collaboration Between Mathematicians in Universiti Putra Malaysia

“…Graphs (Harary, 1969) are important in our daily life with numerous applications as it can be used to portray the relationship between a group of objects. Its applications can be found in many disciplines, including network modelling, information system development analysis and communitybased problem solving (Yow & Luo, 2022).…”