Online Prediction of Switching Graph Labelings with Cluster Specialists

TL;DR

This paper introduces a fast online algorithm for predicting changing graph labelings using cluster specialists, with proven mistake bounds and superior performance on real data.

Contribution

The paper presents a novel cluster specialist-based algorithm for online graph labeling with efficient computation and mistake bounds that adapt to label changes.

Findings

Algorithm achieves $ ext{O}( ext{log} n)$ time per trial.

Mistake bounds vary smoothly with label change magnitude.

Outperforms existing methods on real-world data.

Abstract

We address the problem of predicting the labeling of a graph in an online setting when the labeling is changing over time. We present an algorithm based on a specialist approach; we develop the machinery of cluster specialists which probabilistically exploits the cluster structure in the graph. Our algorithm has two variants, one of which surprisingly only requires $\mathcal{O}(\log n)$ time on any trial $t$ on an $n$-vertex graph, an exponential speed up over existing methods. We prove switching mistake-bound guarantees for both variants of our algorithm. Furthermore these mistake bounds smoothly vary with the magnitude of the change between successive labelings. We perform experiments on Chicago Divvy Bicycle Sharing data and show that our algorithms significantly outperform an existing algorithm (a kernelized Perceptron) as well as several natural benchmarks.