I am an assistant professor at UC San Diego department of Computer Science and Engineering. I am also affiliated with Halıcıoğlu Data Science Institute and Center for Machine-Integrated Computing and Security.

I work on the theory and application of machine learning, especially for large-scale spatiotemporal data. I am generally interested in optimization, deep learning, and spatiotemporal reasoning. I am particularly excited about the interplay between physics and machine learning. My work has been applied to learning dynamical systems in sustainability, health and physical sciences. For more details, see my curriculum vitae. For prospective students, please read this before emailing me.

Contact Info

  • E-mail: roseyu [at] ucsd [dot] edu
  • phone: (858) 246-4724
  • Address: 9500 Gilman Drive, Mail Code 0404, La Jolla, CA 92093

News

Old News

Physics-Guided Deep Learning for Spatiotemporal Dynamics

Incorporate first-principles into deep sequence models

Our physical world is intrinsically spatiotemporal. We design deep learning models that bridge the expressiveness of neural networks and the rich spatiotemporal structures from the data, addressing fundamental challenges of high-dimensionality, high-order correlation, non-linear dynamics and multi-resolution dependencies.

Read More
  • Towards Physics-informed Deep Learning for Turbulent Flow Prediction
    Rui Wang, Karthik Kashinath, Mustafa Mustafa, Adrian Albert, Rose Yu.
    To appear in ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2020
    [Paper] [Code]
  • NAOMI: Non-Autoregressive Multiresolution Sequence Imputation
    Yukai Liu, Rose Yu, Stephan Zheng, Eric Zhan, Yisong Yue.
    Advances in Neural Information Processing Systems (NeurIPS), 2019
    [Paper] [Code]
  • Neural Lander: Stable Drone Landing Control using Learned Dynamics
    Guanya Shi, Xichen Shi, Michael O'Connell, Rose Yu, Kamyar Azizzadenesheli, Anima Anandkumar, Yisong Yue, Soon-Jo Chung
    International Conference on Robotics and Automation (ICRA), 2019
    [Paper] [Demo]

Deep Generative Models for Spatiotemporal Graphs

Structured learning models for spatiotemporal graphs

A graphical model can conveniently encode high-level structures. The nodes of the graph typically represent the problem components, and the edges capture their spatiotemporal interactions. We aim to study a novel deep learning framework with a primary focus on combining graphical models with black-box deep neural networks.

Read More
  • Understanding the Representation Power of Graph Neural Networks in Learning Graph Topology
    Nima Dehmamy, Albert-László Barabási, Rose Yu.
    Advances in Neural Information Processing Systems (NeurIPS), 2019
    [Paper] [Code]
  • A Neural Framework for Learning DAG to DAG Translation
    M. Clara De Paolis Kaluza, Saeed Amizadeh, Rose Yu.
    [Paper]
  • Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting
    Yaguang Li, Rose Yu, Cyrus Shahabi,Yan Liu. International Conference on Learning Representations (ICLR), 2018
    [Paper] [Code]

Interpretable Tensor Models for Spatiotemporal Analysis

High-order tensor models with interpretable factors for domain scientists

Providing interpretable insights for domain scientists is one of the core missions in data-driven decision making. We leverage the efficiency and simplicity of tensor factor models to uncover the hidden spatiotemporal patterns from the data.

Read More
  • Multiresolution Tensor Learning for Efficient and Interpretable Spatial Analysis
    Jung Yeon (John) Park, Kenneth (Theo) Carr, Stephan Zheng, Yisong Yue, Rose Yu
    International Conference on Machine Learning (ICML), 2020
    [Paper] [Code]
  • Learning Tensor Latent Features
    Sung-En Chang, Xun Zheng, Ian E.H. Yen, Pradeep Ravikumar, Rose Yu.
    [Paper]
  • Tensor Regression Meets Gaussian Processes
    Rose Yu, Guangyu Li, Yan Liu. International Conference on Artificial Intelligence and Statistics (AISTATS), 2018
    [Paper]

Preprints

  • Incorporating Symmetry into Deep Dynamics Models for Improved Generalization
    Rui Wang*, Robin Walters*, Rose Yu
    [Paper]
  • Long-term Forecasting using Tensor-Train RNNs
    Rose Yu, Stephan Zheng, Anima Anandkumar, Yisong Yue.
    [Paper] [Code]
  • Socratic Learning: Empowering the Generative Model
    Paroma Varma, Bryan He, Dan Iter,Peng Xu, Rose Yu, Christopher De Sa, Christopher R‌é.
    [Paper]
  • Conferences

  • Multiresolution Tensor Learning for Efficient and Interpretable Spatial Analysis
    Jung Yeon (John) Park, Kenneth (Theo) Carr, Stephan Zheng, Yisong Yue, Rose Yu
    International Conference on Machine Learning (ICML), 2020
    [Paper] [Code]
  • Towards Physics-informed Deep Learning for Turbulent Flow Prediction
    Rui Wang, Karthik Kashinath, Mustafa Mustafa, Adrian Albert, Rose Yu.
    To appear in ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2020
    [Paper] [Code]
  • Understanding the Representation Power of Graph Neural Networks in Learning Graph Topology
    Nima Dehmamy, Albert-László Barabási, Rose Yu.
    Advances in Neural Information Processing Systems (NeurIPS), 2019
    [Paper] [Code]
  • NAOMI: Non-Autoregressive Multiresolution Sequence Imputation
    Yukai Liu, Rose Yu, Stephan Zheng, Eric Zhan, Yisong Yue.
    Advances in Neural Information Processing Systems (NeurIPS), 2019
    [Paper] [Code]
  • Neural Lander: Stable Drone Landing Control using Learned Dynamics
    Guanya Shi, Xichen Shi, Michael O'Connell, Rose Yu, Kamyar Azizzadenesheli, Anima Anandkumar, Yisong Yue, Soon-Jo Chung
    International Conference on Robotics and Automation (ICRA), 2019
    [Paper] [Demo]
  • Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting
    Yaguang Li, Rose Yu, Cyrus Shahabi, Yan Liu.
    International Conference on Learning Representations (ICLR), 2018
    [Paper] [Code]
  • Tensor Regression Meets Gaussian Processes
    Rose Yu, Guangyu Li, Yan Liu.
    International Conference on Artificial Intelligence and Statistics (AISTATS), 2018
    [Paper]
  • Deep Learning: A Generic Approach for Extreme Condition Traffic Forecasting
    Rose Yu, Yaguang Li, Ugur Demiryurek, Cyrus Shahabi, Yan Liu.
    SIAM International Conference on Data Mining (SDM), 2017
    [Paper]
  • Learning from Multiway Data: Simple and Efficient Tensor Regression
    Rose Yu, Yan Liu.
    International Conference on Machine Learning (ICML), 2016
    [Paper] [Slide] [BibTex]
  • Latent Space Model for Road Networks to Predict Time-Varying Traffic
    Dingxiong Deng, Cyrus Shahabi, Ugur Demiryurek, Linhong Zhu, Rose Yu, Yan Liu
    ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2016
    [Paper]
  • Geographic Segmentation via Latent Poisson Factor Model
    Rose Yu, Andrew Gelfand, Suju Rajan, Cyrus Shahabi, Yan Liu.
    ACM International Conference on Web Search and Data Mining (WSDM), 2016
    [Paper] [BibTex]
  • Accelerated Online Low-Rank Tensor Learning for Multivariate Spatio-Temporal Streams
    Rose Yu, Dehua Cheng, Yan Liu.
    International Conference on Machine Learning (ICML), 2015
    [Code] [Paper] [Supplementary] [Video] [Slide] [BibTex]
  • Fast Multivariate Spatio-temporal Analysis via Low Rank Tensor Learning
    Rose Yu*, Mohammad Taha Bahadori*, Yan Liu. (*Equal Contributions)
    Advances in Neural Information Processing Systems (NeurIPS), 2014 Spotlight
    [Code] [Paper] [Supplementary] [BibTex]
  • GLAD: Group Anomaly Detection in Social Media Analysis
    Rose Yu, Xinran He, Yan Liu.
    ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2014
    [Code] [ACM Dataset] [Slide] [Paper-Extended Version] [BibTex]
  • A Feasible Nonconvex Relaxation Approach to Feature Selection
    Cuixia Gao, Naiyan Wang, Qi Yu, Zhihua Zhang.
    AAAI Conference on Artificial Intelligence (AAAI), 2011
    [Paper] [BibTex]

  • Journals

  • Spatio-Temporal Analysis of Social Media Data
    Rose Yu, Yan Liu.
    Encyclopedia of GIS , 2016
  • A Survey on Social Media Anomaly Detection
    Rose Yu, Huida Qiu, Zhen Wen, Ching-Yung Lin, Yan Liu
    ACM SIGKDD Explorations , 2016
    [Paper]
  • GLAD: Group Anomaly Detection in Social Media Analysis (journal version)
    Rose Yu, Xinran He, Yan Liu.
    ACM Transactions on Knowledge Discovery in Data (TKDD), 2015
    [Paper]
  • Current

    PhD Students & Postdocs
    Undergraduates & Master's Students
    • Eliza Huang
    • Xie Fan

    Alumni

    Undergraduates & Master's Students

    CSE 291 (G00) Deep Generative Models

    Fall 2020

    Description: Deep generative models combine the generality of probabilistic reasoning with the scalability of deep learning. This research area is at the forefront of deep learning and has given state-of-the-art results in text generation, video synthesis, molecular design, amongst many others. This course will cover recent advances in deep generative models, including variational autoencoders, generative adversarial networks, autoregressive models, and normalizing flow models. This is a Ph.D. level course with emphasis on mathematical principles as well as practical know-how. The course will be a combination of lectures, student presentations, and team projects.

    Courses Taught at Northeastern University

    CS 7140 Advanced Machine Learning, Spring 2020

  • Course Website
  • CS 3950/4950 Introduction to Computer Science Research, Fall 2019

  • 3950 Course Website    4950 Course Website
  • CS 7180 Special Topics in AI: Deep Learning, Spring 2019

  • Course Website
  • CS 6140 Machine Learning, Fall 2018

  • Course Website    Course Syllabus