Summary of Pagerank Bandits For Link Prediction, by Yikun Ban et al.
PageRank Bandits for Link Prediction
by Yikun Ban, Jiaru Zou, Zihao Li, Yunzhe Qi, Dongqi Fu, Jian Kang, Hanghang Tong, Jingrui He
First submitted to arxiv on: 3 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A machine learning educator may summarize this abstract by stating that researchers have proposed various methods for link prediction in graph learning, including those based on similarity metrics and Graph Neural Networks. However, these approaches often rely on conventional supervised learning, which can be limiting when trying to adapt to changing customer interests or address the dilemma of exploitation versus exploration in link prediction. To address these challenges, this paper reformulates link prediction as a sequential decision-making process and proposes a novel fusion algorithm called PRB (PageRank Bandits) that combines contextual bandits with PageRank for collaborative exploitation and exploration. The authors also introduce a new reward formulation and provide theoretical performance guarantees for PRB. The algorithm is evaluated in both online and offline settings, comparing it to bandit-based and graph-based methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary For curious learners or general audiences, this paper is about finding connections between things on the internet. Imagine trying to predict which movies someone might like based on what they’ve watched before. Researchers have been working on ways to make these predictions more accurate, but there’s a problem: the world is constantly changing, and people’s tastes can shift quickly. To solve this challenge, scientists have developed a new way of thinking about link prediction as a series of decisions made one at a time. They’ve also created an algorithm called PRB that combines different techniques to make better predictions. |
Keywords
» Artificial intelligence » Machine learning » Supervised
