Shen Yan

I am a Research Scientist at Meta. Prior to Meta, I received my Ph.D. degree in Computer Science at University of Southern California in 2022, where I was fortunate and honored to have Dr. Emilio Ferrara as my advisor and worked as a Research Assistant at Information Sciences Institute. I got my Master’s Degree in Computer Science at University of Chinese Academy of Sciences in 2017.

My research interests lie at the intersections of Human Behavior Understanding, Network Science, and AI Fairness, and I am particularly interested in understanding the heterogeneity of human behaviors, and designing responsible AI systems. I am broadly interested in Computational Social Science.

Fairness-Aware Machine Learning for Multimodal Human Behavior Data

Behavioral modeling has found broad applicabilitiy in many decision-making domains, including health and job performance evaluation, financial and employment scrutiny, etc. Hence, guaranteeing model fairness in behavioral modeling is an open challenge for real-world applications. We propose and develop machine learning models that will identify, account for, and mitigate the biases of multimodal human behavior modeling methods.

Relevant publications: CIKM’20, ICMI’20, ACII’21, AAAI’23

Tracking Individual pErformance with Sensors (TILES)

TILES is a project focused on the analysis of stress, task performance, behavior, and other factors pertaining to professionals engaged in a high-stress workplace environment. Biological, environmental, and contextual data is collected from hospital nurses and staff both in the workplace and at home over a period of 10 weeks. Labels of human experience are collected using a variety of psychologically validated questionnaires sampled on a daily basis at different times during the day.

We design machine learning models to estimate human behaviors from the sensory data, which are aiming to provide promising solutions to generalizable, trust, and fair mobile health systems.

Relevant publications: BESC’18, ICHI’19, JHIR, IMWUT

See full list of publications here. [Google Scholar]

AI Fairness

• FairFed: Enabling Group Fairness in Federated Learning
  Y.-H. Ezzeldin*, S. Yan*, C. He, E. Ferrara, S. Avestimehr
  37th AAAI Conference on Artificial Intelligence (AAAI'23)
  NeurIPS 2021 Workshop on Federated Learning: Privacy, Fairness, Robustness, Personalization and Data Ownership

• Mitigating the Bias of Heterogeneous Human Behavior in Affective Computing
  S. Yan, H.-T Kao, K. Lerman, S. Narayanan, and E. Ferrara
  9th International Conference on Affective Computing & Intelligent Interaction (ACII'21)

• Fair Class Balancing: Enhancing Model Fairness without Observing Sensitive Attributes [code]
  S. Yan, H.-T Kao, and E. Ferrara
  29th ACM International Conference on Information and Knowledge Management (CIKM'20)

• Mitigating Biases in Multimodal Personality Assessment [code]
  S. Yan, D. Huang, and M. Soleymani
  22nd ACM International Conference on Multimodal Interaction (ICMI'20)

Human Behavior Understanding & Healthcare Informatics

• User-Based Collaborative Filtering Mobile Health System
  H.-T Kao, S. Yan, H. Hosseinmardi, S. Narayanan, K. Lerman, and E. Ferrara
  Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IWMUT)

• Estimating Individualized Daily Self-Reported Affect with Wearable Sensors
  S. Yan, H. Hosseinmardi, H.-T Kao, S. Narayanan, K. Lerman, and E. Ferrara
  7th IEEE International Conference on Healthcare Informatics (ICHI'19)

Social Networks

• What Does Perception Bias on Social Networks Tell Us About Friend Count Satisfaction? [code]
  S. Yan, K.M. Altenburger, Y.-C. Wang, J. Cheng
  The ACM Web Conference 2022

• Political polarization drives online conversations about COVID-19 in the United States
  J. Jiang, E. Chen, S. Yan, K. Lerman, and E. Ferrara
  Human Behavior and Emerging Technologies

• Understanding Cyberbullying on Instagram and Ask.fm via Social Role Detection
  H.-T Kao, S. Yan, D. Huang, N. Bartley, H. Hosseinmardi, and E. Ferrara
  4th Workshop on Computational Methods in Online Misbehavior Co-located with The Web Conference (CyberSafety'19)

Data Privacy

• DynaEgo: Privacy-preserving collaborative filtering recommender system based on social-aware differential privacy
  S. Yan, S. Pan, W.-T. Zhu, and K. Chen
  18th International Conference on Information and Communication Security (ICICS'16)

• Towards privacy-preserving data mining in online social networks: Distance-grained and item-grained differential privacy
  S. Yan, S. Pan, Y. Zhao, and W.-T. Zhu
  21st Australasian Conference on Information Security and Privacy (ACISP'16)