Understanding the limitations of network online learning

Abstract

Studies of networked phenomena, such as interactions in online social media, often rely on incomplete data, either because these phenomena are partially observed, or because the data is too large or expensive to acquire all at once. Analysis of incomplete data leads to skewed or misleading results. In this paper, we investigate limitations of learning to complete partially observed networks via node querying. Concretely, we study the following problem: given (i) a partially observed network, (ii) the ability to query nodes for their connections (e.g., by accessing an API), and (iii) a budget on the number of such queries, sequentially learn which nodes to query in order to maximally increase observability. We call this querying process Network Online Learning and present a family of algorithms called NOL*. These algorithms learn to choose which partially observed node to query next based on a parameterized model that is trained online through a process of exploration and exploitation. Extensive experiments on both synthetic and real world networks show that (i) it is possible to sequentially learn to choose which nodes are best to query in a network and (ii) some macroscopic properties of networks, such as the degree distribution and modular structure, impact the potential for learning and the optimal amount of random exploration.

HypOp: Distributed Constrained Combinatorial Optimization leveraging Hypergraph Neural Networks

“…Our goal is to show the limitations of PI-GNN on some graphs. This analysis is inspired by the results of [25], where the authors study the limitations of the network online learning algorithms by considering different types of graphs.…”

Identifying Communities with Modularity Metric Using Louvain and Leiden Algorithms

“…For the performance testing of each community detection algorithm, it is preferable to run on networks that can be modified regarding the mesoscopic characteristics, including the distribution of degrees, the extent of local clustering, and the modularity of the global structure (LaRock et al, 2020). First, the degree distribution is important in determining whether learning is feasible or beneficial.…”

Vulnerability and robustness in artificial intelligence for complex networks

“…For example, PATHATTACK assumes full knowledge of the graph, which in many settings is not realistic. Combining a method like PATHATTACK with a method to explore more of an unseen network, like Network Online Learning* [84] may provide a more apt formulation for such scenarios. There are many interesting theoretical questions raised by this work.…”