Simulation has emerged as a popular method to study the long-term societal consequences of recommender systems. This approach allows researchers to specify their theoretical model explicitly and observe the evolution of system-level outcomes over time. However, performing simulation-based studies often requires researchers to build their own simulation environments from the ground up, which creates a high barrier to entry, introduces room for implementation error, and makes it difficult to disentangle whether observed outcomes are due to the model or the implementation.
We introduce T-RECS, an open-sourced Python package designed for researchers to simulate recommendation systems and other types of sociotechnical systems in which an algorithm mediates the interactions between multiple stakeholders, such as users and content creators. To demonstrate the flexibility of T-RECS, we perform a replication of two prior simulation-based research on sociotechnical systems. We additionally show how T-RECS can be used to generate novel insights with minimal overhead. Our tool promotes reproducibility in this area of research, provides a unified language for simulating sociotechnical systems, and removes the friction of implementing simulations from scratch.
A Review of Modern Recommender Systems Using Generative Models (Gen-RecSys)
Polytechnic University of Bari(巴里理工大学);University of California(加利福尼亚大学);University of Toronto(多伦多大学);Amazon(亚马逊公司);Bespoke Labs;University of Edinburgh(爱丁堡大学);University of Exeter(埃克塞特大学);LMU Munich(慕尼黑大学)
Towards Recommender Systems LLMs Playground (RecSysLLMsP): Exploring Polarization and Engagement in Simulated Social Networks
Institute for Artificial Intelligence Research and Development of Serbia(塞尔维亚人工智能研究与发展研究所);University of Belgrade(贝尔格莱德大学);Polytechnic University of Bari(巴里理工大学);University of Kragujevac(克拉古耶瓦茨大学)
ARTAI: An Evaluation Platform to Assess Societal Risk of Recommender Algorithms
University College Dublin(都柏林大学学院);School of Computer Science(计算机科学学院);School of Information and Communication(信息与通信学院);Dublin City University(都柏林城市大学);School of Computing(计算机学院);University College Cork(科克大学学院);School of Computer Science & IT(计算机科学与信息技术学院);Insight SFI Research Centre for Data Analytics(Insight SFI数据分析研究中心)
RecSys Challenge 2024: Balancing Accuracy and Editorial Values in News Recommendations
JP/Politikens Media Group(JP/Politikens媒体集团);Technical University of Denmark(丹麦技术大学);Copenhagen Business School(哥本哈根商学院);ETH Zürich(苏黎世联邦理工学院);University of Bari Aldo Moro(巴里大学);Politecnico di Bari(巴里理工大学);Indian Institute of Management Visakhapatnam(维萨卡帕特南印度管理学院)
Toward Holistic Evaluation of Recommender Systems Powered by Generative Models
Polytechnic University of Bari(巴里理工大学);Duke University(杜克大学);Bespoke Labs(Bespoke实验室);ETH Zurich(苏黎世联邦理工学院);Autónoma University of Madrid(马德里自治大学);University of California, San Diego(加州大学圣地亚哥分校)
RGRecSys: A Toolkit for Robustness Evaluation of Recommender Systems
University of Illinois at Chicago(伊利诺伊大学芝加哥分校);Salesforce Research(Salesforce研究院);Rutgers University(罗格斯大学)
健壮的机器学习是一个越来越重要的主题,它专注于开发对各种形式的不完美数据有弹性的模型。由于推荐系统在在线技术中的普遍性,研究人员已经进行了一些鲁棒性研究,主要集中在数据稀疏性和配置文件注入攻击。相反,我们提出了一个更全面的观点,推荐系统的鲁棒性,包括多个维度-鲁棒性相对于子种群,转换,分布差异,攻击和数据稀疏。虽然有几个库允许用户比较不同的推荐系统模型,但没有软件库用于在不同场景下对推荐系统模型进行全面的鲁棒性评估。作为我们的主要贡献,我们提出了一个鲁棒性评估工具包,Robustness Gym for RecSys(RGRecSys --https://www.github.com/salesforce/RGRecSys),它允许我们快速和统一地评估推荐系统模型的鲁棒性。
Insight SFI Research Centre for Data Analytics(Insight SFI数据分析研究中心);School of Information and Communication Studies(信息与传播研究学院);University College Dublin(都柏林大学)