September 21-26, 2021, All over the world (virtual)

Combining Physical and Data-Driven Knowledge in Ubiquitous Computing

Ubicomp 2021 Workshop


In the real-world ubiquitous computing systems, it is difficult to require a significant amount of data to obtain accurate information through pure data-driven methods. The performance of data-driven methods relies on the quantity and ‘quality’ of data. They perform well when a sufficient amount of data is available, which is regarded as ideal conditions. However, in real-world systems, collecting data can be costly or impossible due to practical limitations. On the other hand, it is promising to utilize physical knowledge to alleviate these issues of data limitation. The physical knowledge includes domain knowledge from experts, heuristics from experiences, analytic models of the physical phenomena and etc. We aim to bring researchers that explore this direction together and search for systematic solutions across various applications.


The goal of the workshop is to explore the intersection between (and the combination of) data and physical knowledge. The workshop aims to bring together domain experts that explore the physical understanding of the data, practitioners that develop systems and the researchers in traditional data-driven domains. The workshop welcomes papers, which focuses on addressing these issues in different applications/domains as well as algorithmic and systematic approaches to applying physical knowledge. Therefore, we further seek to develop a community that systematically analyzes the data quality regarding inference and evaluates the improvements from physical knowledge. Preliminary and on-going work is welcomed.

  • The accepted papers are invited to be extended for a journal fast-track submission at (1) Digital Signal Processing: A Review Journal; and (2) Intelligent and Converged Networks with at least 30% contribution on novelty
  • Topics of Interests

    Topics of interests include, but are not limited to, the follows:
    • - Innovations in learning algorithms that combine physical knowledge or models for sensor perception and understanding
    • - Experiences, challenges, analysis, and comparisons of sensor data in terms of its physical properties
    • - Sensor data processing to improve learning accuracy
    • - Machine learning and deep learning with physical knowledge on sensor data
    • - Mobile and pervasive systems that utilize physical knowledge to enhance data acquisition
    • - System services such as time and location estimation enhanced by additional physical knowledge
    • - Heterogeneous collaborative sensing based on physical rules
    • - Distributed sensing for cyber-physical systems
    • - Advanced machine learning algorithms and solutions for efficient sensing
    The application areas include but not limited to:
    • - Human-centric sensing applications
    • - Environmental and structural monitoring
    • - Smart cities and urban health
    • - Health, wellness and medical
    • - Smart energy systems and intelligent transportation networks

    Successful submissions will explain why the topic is relevant to the data limitation caused problem that may be solved through the physical understanding of domain knowledge. In addition to citing relevant, published work, authors must cite and relate their submissions to relevant prior publications of their own. Ethical approval for experiments with human subjects should be demonstrated as part of the submission.

    Important Dates

    Submission Deadline: June 29, 2021

    Notifications: July 15, 2021

    Camera-ready: July 31, 2021

    Workshop: September 27, 2021

    Submission Guidelines

    Submissions can be made at PCS. The workshop paper for CPD can be submitted via: SIGCHI -> UbiComp/ISWC 2021 -> UbiComp/ISWC 2021 Workshop: CPD

    Please submit short papers that are at most 5 single-spaced 8.5” x 11” pages, including figures and tables, but excluding references, two-column format, using 10-point type on 11-point (tight single-spaced) leading, with a maximum text block of 7” wide x 9” deep with an inter-column spacing of .25”. Submissions may include as many pages as needed for references.

    ACM Template can be found here.

    The accepted paper will be published in the proceedings with Ubicomp papers this year.


    Workshop Chairs

    Wenbo Ding Tsinghua-Berkeley Shenzhen Institute

    Chenshu Wu University of Maryland

    Weitao Xu City University of Hong Kong

    Advising Committee

    Pei Zhang University of Michigan

    Hae Young Noh Stanford University

    Jie Liu IEEE Fellow Harbin Institute of Technology

    Ercan Engin Kuruoglu National Research Council of Italy

    Workshop TPC Chairs

    Xinlei Chen Tsinghua University

    Shijia Pan University of California Merced

    Sicong Liu Xiamen University

    Technical Programm Committee

    Amir H. Alavi University of Pittsburgh

    Heba Aly Amazon

    Ishara Dharmasena Loughborough University, UK

    Di Jin Amazon Alexa AI

    Abdelwahed Khamis University of New South Wales

    Guohao Lan Duke University

    Chengwen Luo Shenzhen University

    Joao Palotti Qatar Computing Research Institute, HBKU

    Parth Pathak George Mason University

    Longfei Shangguan Microsoft Cognition

    Stephan Sigg Aalto University

    Shuai Wang Southeast University

    Bo Wei Northumbria University

    Yaxiong Xie Princeton University

    Susu Xu Qualcomm AI Research

    Fengli Xu University of Chicago

    Wanli Xue University of New South Wales

    Zimu Zhou Singapore Management University

    Web Chair

    Zihan Wang Tsinghua-Berkeley Shenzhen Institute

    Publicity Chair

    Le Liang Southeast University


    27th September 2021 (UTC+8)

    Keynote Speech (9:00-9:40)

    Efficient On-Device Deep Learning: towards Ubiquitous Intelligence on the Edge

    Speaker: Prof. Yunxin Liu, Tsinghua University

    Abstract: With the advances of hardware, software, and artificial intelligence (AI), there is a new computing paradigm shift from centralized intelligence in the cloud to distributed intelligence on the edge. In the era of edge computing, it is critical to infuse AI to empower diverse edge devices and applications. This talk overviews the challenges and opportunities of on-device deep learning and introduces our recent research work on making on-device deep-learning more efficient, focusing on how to build affordable AI models customized for diverse edge devices and how to maximize the performance of on-device model inference by fully utilizing the heterogeneous computing resources.

    Oral Session 1 (9:40-11:40)

    All presenters need to enter the online conference room at least 10 minutes in advance, and ensure all the audio and video devices are working properly.

    9:40-10:00 Electromagnetic Vibration Tactile Feedback for Biological and Artificial Wave Signals

    Presenter: Xiaosa Li, Wuhan University

    10:00-10:20 Understanding Structural Hole Spanners in Location-Based Social Networks: A Data-Driven Study

    Presenter: Xiaoxin He, Fudan University

    10:20-10:40 Quantity or Quality: Data Enabled Online Energy Dispatch

    Presenter: Jingshi Cui, Tsinghua University

    10:40-11:00 MassHog: Weight-Sensitive Occupant Monitoring for Pig Pens using Actuated Structural Vibrations

    Presenter: Jesse R Codling, University of Michigan

    11:00-11:20 PIWIMS: Physics Informed Warehouse Inventory Monitory via Synthetic Data Generation

    Presenter: Prabh Simran Singh Baweja, Carnegie Mellon University

    11:20-11:40 TACNet: Task-aware Electroencephalogram Classification for Brain-Computer Interface through A Novel Temporal Attention Convolutional Network

    Presenter: Xiaolin Liu, Beihang University

    Oral Session 2 (13:00-17:20)

    All presenters need to enter the online conference room at least 10 minutes in advance, and ensure all the audio and video devices are working properly.

    13:00-13:20 Deep Learning Based Underwater Acoustic Channel Estimation Exploiting Physical Knowledge on Channel Sparsity

    Presenter: Longjie Gao, Xiamen University

    13:20-13:40 Physical Knowledge Driven Multi-scale Temporal Receptive Field Network for Compressed Video Action Recognition

    Presenter: Lijun He, Xi'an Jiaotong University

    13:40-14:00 Three-Dimensional Indoor Visible Light Localization: A Learning-Based Approach

    Presenter: Danping Su, Xiamen University

    14:00-14:20 Blind Calibration by Maximizing Correlation

    Presenter: Guodong Li, Tsinghua University

    14:20-14:40 Generative Adversarial Network Enabled Sparse Signal Compression and Recovery for Internet of Medical Things

    Presenter: Tiankuo Wei, Xiamen University

    14:40-15:00 A Simple and Fast Human Activity Recognition System Using Radio Frequency Energy Harvesting

    Presenter: Tao Ni, City University of Hong Kong

    15:00-15:20 Few-Shot Cross Domain Battery Capacity Estimation

    Presenter: Zihao Zhou, Tsinghua-Berkeley Shenzhen Institute

    15:20-15:40 Mobility Data-driven Complete Dispatch Framework for the Ride-hailing Platform

    Presenter: Jiaman Wu, Tsinghua University

    15:40-16:00 Dark-Channel Mixed Attention Based Neural Networks for Smoke Detection in Fog Environment

    Presenter: Le Yang, Xi’an Jiaotong University

    16:00-16:20 Data-driven Clustering in Ad-hoc Networks based on Community Detection

    Presenter: Shufan Huang, Shanghai Jiao Tong University

    16:20-16:40 HTPad: Hexagon-fractal TENG Pad for Scalable Touch Control

    Presenter: Xu Yang, Tsinghua University

    16:40-17:00 TIP-Air: Tracking Pollution Transfer for Accurate Air Quality Prediction

    Presenter: Yun Cheng, ETH Zurich

    17:00-17:20 TriboGait: A deep learning enabled triboelectric gait sensor system for human activity recognition and individual identification

    Presenter: Jiarong Li, Tsinghua University


    The CPD 2021 workshop is part of Ubicomp 2021.

    Online Conference Room (Main)


    Room number : 917 451 4654

    Password : 202021

    Click to add to your conference list

    Backup Conference Room

    Only activated when ZOOM is not available

    VOOV/Tencent Meeting(VooV)
    Room number : 638 820 384
    Click to add to your conference list