Yu Haibin

Currently I am a machine learning engineer at TikTok . My work is on TikTok live strategy recommendation. Previously, I was a machine learning engineering at Tencent , where I took charge in the recommendation algorithms and strategies for Tencent’s advertisement through modeling technologies including deep learning, representation learning, multi-task learning, causal inference, and sequence modeling. My research expertise lies in probabilistic machine learning, optimization, generative modeling, approximate inference and recommendation systems.

Email  /  Google Scholar  /  Github

• May 2024: Our paper "Ads Recommendation in a Collapsed and Entangled World" is accepted to KDD 2024!

• May 2024: Invited to serve as a reviewer for NeurIPS 2024!

• Dec 2022: Our paper "Recursive Reasoning-Based Training-Time Adversarial Machine Learning" is accepted to Artificial Intelligence!

• Nov 2022: Our paper "AdaTask: A Task-aware Adaptive Learning Rate Approach to Multi-task Learning" is accepted to AAAI 2023!

• May 2022: Our paper "On Provably Robust Meta-Bayesian Optimization" is accepted to UAI 2022!

• Mar 2022: Invited to serve as a reviewer for NeurIPS 2022

• Nov 2021: Invited to serve as a reviewer for ICML 2022

• Massachusetts Institute of Technology (MIT)   (Jan 2017 - Aug 2017)
  • Visiting Scholar
• National University of Singapore   (Aug 2014 - Jan 2020)
  • Ph.D. student in Artificial Intelligence, Department of Computer Science
  • Advisors: Bryan Kian Hsiang Low (NUS) & Patrick Jaillet (MIT)
  • Supported by Singapore-MIT Alliance for Research and Technology (SMART) Graduate Fellowship
• Beihang University   (Sept 2010 - Jun 2014)
  • B.Eng. in Mechanical Engineering

• Disentangled Representation with Cross Experts Covariance Loss for Multi-Domain Recommendation
  Zhutian Lin, Junwei Pan, Haibin Yu, Xi Xiao, Ximei Wang, Zhixiang Feng, Shifeng Wen, Shudong Huang, Lei Xiao
  Pre-print, 2024

• Ads Recommendation in a Collapsed and Entangled World
  Junwei Pan, Wei Xue, Ximei Wang, Haibin Yu , Xun Liu, Shijie quan, Xueming Qiu, Dapeng Liu, Lei Xiao and Jie Jiang
  In Proceedings of 2024 International Conference on Knowledge Discovery and Data Mining (KDD-24, Applied Data Science Track)
  Acceptance rate: 20%.

• Genetic Variation and Nonalcoholic Fatty Liver Disease:Field Synopsis, Systematic Meta-Analysis, and Epidemiological Evidence
  Yamei Li, Xiang Xiao, Jie Wang, Yixu Liu, Xiongfeng Pan, Haibin Yu, Jiayou Luo and Miyang Luo
  In Biomedical and Environmental Sciences

• Recursive Reasoning-Based Training-Time Adversarial Machine Learning
  Yizhou Chen, Zhongxiang Dai, Haibin Yu, Kian Hsiang Low and Teck-Hua Ho
  In Artificial Intelligence (Special Issue on Risk-Aware Autonomous Systems: Theory and Practice), volume 315, article 103837

• AdaTask: A Task-aware Adaptive Learning Rate Approach to Multi-task Learning
  Enneng Yang, Junwei Pan, Ximei Wang, Haibin Yu, Li Shen, Xihua Chen, Lei Xiao, Jie Jiang and Guibing Guo
  In Proceedings of the 37th AAAI Conference on Artificial Intelligence (AAAI-23)
  Acceptance rate: 19.6%.

• On Provably Robust Meta-Bayesian Optimization
  Zhongxiang Dai, Yizhou Chen, Haibin Yu, Kian Hsiang Low and Patrick Jaillet
  In Proceedings of the 38th Conference on Uncertainty in Artificial Intelligence, 2022 (UAI-22)
  Acceptance rate: 32.3%.

• Convolutional Normalizing Flows for Deep Gaussian Processes
  Haibin Yu, Kian Hsiang Low, Patrick Jaillet and Dapeng Liu
  In Proceedings of the Internatioanl Joint Conference of Neural Networks, 2021 (IJCNN-21)
  Acceptance rate: 59.3%.

• Implicit Posterior Variational Inference for Deep Gaussian Process
  Haibin Yu*, Yizhou Chen*, Kian Hsiang Low and Patrick Jaillet
  In Proceedings of the 33rd Conference on Neural Information Processing Systems, 2019 (NeurIPS-19)
  Acceptance rate: 3% (spotlight).

• Bayesian Optimization Meets Bayesian Optimal Stopping
  Zhongxiang Dai, Haibin Yu, Kian Hsiang Low and Patrick Jaillet.
  In Proceedings of the 36th Internatioanl Conference of Machine Learning, 2019 (ICML-19)
  Acceptance rate: 22.6%.

• Stochastic Variational Inference for Bayesian Sparse Gaussian Process Regression
  Haibin Yu, Trong Nghia Hoang, Kian Hsiang Low and Patrick Jaillet
  In Proceedings of the Internatioanl Joint Conference of Neural Networks, 2019 (IJCNN-19)
  Acceptance rate: 52.4%.

• Outstanding Graduate of Beihang University,    2014

• Singapore-MIT Alliance for Research and Technology (SMART) Graduate Fellowship,    2014-2018

• JDDiscovery Population Dynamics Census and Prediction Competition 2018 (annual competition hosted by JD.com): global champion, ranked 1st among > 2,100 teams, Jan 2019 (News in English, News in Chinese)

• Invited reviewer for IEEE Transactions on Cybernatics.
• Invited reviewer for Annual Conference on Neural Information Processing Systems (NeurIPS 2020-2024).
• Invited reviewer for International Joint Conference on Artificial Intelligence (IJCAI 2020-2021).
• Invited reviewer for International Conference on Learning Representations (ICLR 2021-2022).
• Invited reviewer for International Conference on Machine Learning (ICML 2022).

• Why Probabilistic Machine Learning Comes to Rescue,    2019.10
  • Wilmar@NUS Lab, NUS, Singapore
• Implicit Posterior Variational Inference for Deep Gaussian Processes,    2019.11
  • AI Seminar, NUS, Singapore
• Bayesian Machine Learning and Automatic Machine Learning Come to Rescue,    2019.12
  • Sun Yat-sen University Forum for International Young Scholars, Guangzhou, China
• Bayesian Machine Learning and Automatic Machine Learning Come to Rescue,    2020.06
  • Soochow University Forum for International Young Scholars, Suzhou, China