# Anomaly Detection in Networked Bandits

## TL;DR

This paper introduces a novel bandit algorithm that leverages network information to personalize recommendations and detect anomalies in social networks, with proven regret bounds and experimental validation.

## Contribution

It presents a new network-aware bandit algorithm that simultaneously learns user preferences and detects anomalies, advancing online learning in social network analysis.

## Key findings

- Proven upper bound on the algorithm's regret.
- Effective anomaly detection demonstrated on real-world data.
- Outperforms existing collaborative bandit algorithms in experiments.

## Abstract

The nodes' interconnections on a social network often reflect their dependencies and information-sharing behaviors. Nevertheless, abnormal nodes, which significantly deviate from most of the network concerning patterns or behaviors, can lead to grave consequences. Therefore, it is imperative to design efficient online learning algorithms that robustly learn users' preferences while simultaneously detecting anomalies. We introduce a novel bandit algorithm to address this problem. Through network knowledge, the method characterizes the users' preferences and residuals of feature information. By learning and analyzing these preferences and residuals, it develops a personalized recommendation strategy for each user and simultaneously detects anomalies. We rigorously prove an upper bound on the regret of the proposed algorithm and experimentally compare it with several state-of-the-art collaborative contextual bandit algorithms on both synthetic and real-world datasets.