Yanjie Fu

AI systems, unlike humans, are brittle, not robust, often struggle when faced with novel situations, and highly sensitive to small perturbations, which can lead to catastrophically poor performance. My research aims to develop trusted and safe machine intelligence, by building tools to address learning framework, algorithmic, data, and computing challenges. My perspective is to connect computing issues in representation learning (imperfect data, structure knowledge), self-supervised learning (limitation of labels), interactive learning (weak supervision and uncertain environments), adaptive learning (shift and drifted environment), stream learning (limitation of memory) and more as disruption-robust learning. This connection not only provides a unified understanding, but also paves a principled and innovative way to design trusted and safe systems as a disruption-robust framework. I execute two important steps steps towards this vision. The first step (data representation construct) aims to integrate structure knowledge, self-optimization, explainability to achieve deep robust representation to fight imperfect and complexity data. The second step (learning strategy construct) aims to integrate robust representations with adaptive and interactive learning to fight uncertain and constrained environments.

Ph.D. students graduated from my research group:

• Geospatial Generative AI for Automated Urban Planning and Urban Informatics
  • Ph.D. Dissertation Student: Dongjie Wang, an incoming tenure-track assistant professor at the University of Kansas
  • Reimagining city configuration: Automated urban planning via adversarial learning (SIGSPatial best paper runner up)Slides
  • Human-instructed Deep Hierarchical Generative Learning for Automated Urban Planning (AAAI)Slides
  • Automated Urban Planning for Reimagining City Configuration via Adversarial Learning: Quantification, Generation, and Evaluation (ACM Trans on SAS): Slides
• Self-optmizing Feature Selection
  • Joint Exchange Program Advisee: Meng Xiao, an assistant professor at Chinese Academy of Sciences
  • Traceable group-wise self-optimizing feature transformation learning: A dual optimization perspective(TKDD)Slides
  • Beyond Discrete Selection: Continuous Embedding Space Optimization for Generative Feature Selection (IEEE ICDM): Slides
• Deep Graph Learning 
  • Joint Exchange Program Advisee: Ziyue Qiao, an assistant professor at the Great Bay University
  • A dual-channel semi-supervised learning framework on graphs via knowledge transfer and meta-learning (ACM TWEB)Slides
• Deep Time Series Learning
  • Ph.D. Dissertation Student: Wei Fan (2020-2023), a postdoc at University of Oxford
  • DEPTS: Deep Expansion Learning for Periodic Time Series Forecasting. (ICLR22): Slides
  • Dish-TS: A General Paradigm for Alleviating Distribution Shift in Time Series Forecasting. (AAAI23): Slides
  • IN-Flow: Address Distribution Shift in Time Series Forecasting. (NeurIPS submission): Slides
• AI for Smart Education and Learning Science
  • Joint Exchange Program Advisee: Lu Jiang, an assistant professor at Dalian Maritime University
  • Reinforced explainable knowledge concept recommendation in MOOCs (ACM TIST): Slides
• Reinforcement Learning for Automated Data Science
  • Ph.D. Dissertation Student: Kunpeng Liu(2017-2022), a tenure-track assistant professor at Portland State University
  • Automated Feature Selection: A Reinforcement Learning Perspective (TKDE): Slides
  • Interactive Reinforcement Learning for Feature selection with Decision Tree in the Loop. (TKDE): Slides
  • Efficient Reinforced Feature Selection via Early Stopping Traverse Strategy. (ICDM Best Finalist): Slides
  • Automating Feature Subspace Exploration via Multi-Agent Reinforcement Learning. (KDD): Slides
• Structure Knowledge Guided Spatial-Temporal Graph and Knowledge Graph Representation Learning: Summary
  • Ph.D. Dissertation Student: Pengyang Wang (2017-2021), a tenure-track assistant professor at University of Macau
  • Mutual Information Based Substructured Representation Learning (IJCAI20): Slides
  • Adversarial Substructured Representation Learning (KDD19): Slides
  • Peer and Temporal-Aware Representation Learning (KDD'18 Best Student Paper Finalist): Slides
  • Collective Embedding with Periodic Spatial-temporal Mobility Graphs (TIST): Slides
• Machine Learning for Human Mobility Modeling: Summary
  • Joint Exchange Program Advisee: Pengfei Wang, an associate professor at Chinese Academy of Sciences
  • Representing Urban Forms: A Collective Learning Model with Heterogeneous Human Mobility Data(TKDE) : Slides
  • Human Mobility Synchronization and Trip Purpose Detection with Mixture of Hawkes Processes(KDD'17) : Slides
  • Spotting Trip Purposes from Taxi Trajectories: A General Probabilistic Model (TIST) : Slides