October 12, 2018, Singapore
ABOUT THE WORKSHOP
Real-world ubiquitous computing systems face the challenge of requiring a significant amount of data to obtain accurate information through pure data-driven approaches. The performance of these data-driven systems greatly depends on the quantity and `quality' of data. In ideal conditions, pure data-driven methods perform well due to the abundance of data. However, in real-world systems, collecting data can be costly or impossible due to practical limitations. Physical knowledge, on the other hand, can be used to alleviate these issues of data limitation. This physical knowledge can include domain knowledge from experts, heuristics from experiences, as well as analytic models of the physical phenomena.
This workshop aims to explore the intersection between (and the combination of) data and physical knowledge. The workshop will 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 addressing these issues in different applications/domains as well as algorithmic and systematic approaches to apply physical knowledge. Therefore, we further seek to develop a community that systematically analyzes the data quality regarding inference and evaluates the improvements from the physical knowledge. Preliminary and on-going work are welcomed.
CALL FOR PAPER
With the physical knowledge, we can infer the target information 1) more accurately compared to the pure data-driven model, or 2) with limited (labeled) data, since it is often difficult to obtain a large amount of (labeled) data under various conditions. In recent years, researchers combine this physical knowledge with traditional data-driven approaches to improve the computing performance with limited (labeled) data. We aim to bring researchers that explore this direction together and search for systematic solutions across various applications. 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
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
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.
Submission Guidelines
Please submit short papers using the SIGCHI Extended Abstract format with no more than 8 pages of content. Submissions may include as many pages as needed for references. The submissions should not be anonymous.
Format in GitHub: here
Template download: here
Submission site: link
ORGANIZERS
Advising Committee
Pei Zhang Carnegie Mellon University
Hae Young Noh Carnegie Mellon University
Jie Liu Microsoft AI & Research
Technical Programm Committee
Josiah Hester Northwestern University
Mi Zhang Michigan State University
Tam Vu University of Colorado Boulder
Yuan Tian University of Virginia
Jun Han National University of Singapore
Kent Lyons Tesla
Yong Li Tsinghua University
Yue Wang AcuSys, Inc
Pan Hu Stanford University
Mo Li Nanyang Technological University
Kaifei Chen UC Berkeley
Bing Liu Carnegie Mellon University
AGENDA
Registration opens (8:00-9:00)
Welcome! (9:00-9:30)
Speaker: Pei Zhang
Session 1: System Sensibility (9:30-10:30), Chair: Jun Han (NUS)
Volatile Organic Compounds Recognition Using a Smartphone Camera and Fluorometric Sensors.
Jungmo Ahn, Hyungi Kim, Eunha Kim, JeongGil Ko (Ajou University)
Moisture Based Perspiration Level Estimation.
Ji Jia, Chentian Xu (Columbia University) Shijia Pan (Carnegie Mellon University) Stephen Xia, Peter Wei (Columbia University) Hae Young Noh, Pei zhang (Carnegie Mellon University) Xiaofan Jiang (Columbia University)
Human Gait Monitoring Using Footstep-Induced Floor Vibrations Across Different Structures.
Mostafa Mirshekari, Jonathon Fagert, Amelie Bonde, Pei Zhang, Hae Young Noh (Carnegie Mellon University)
Coffee break (10:30-11:00)
Session 2: Understand and Fuse Data (11:00-12:00), Chair: Tarek Abdelzaher (UIUC)
Lightweight Modeling of User Context Combining Physical and Virtual Sensor Data.
Mattia Giovanni Campana (IIT-CNR) Dimitris Chatzopoulos (HKUST) Franca Delmastro (IIT-CNR) Pan Hui (HKUST and University of Helsinki)
PGA: Physics Guided and Adaptive Approach for Mobile Fine-Grained Air Pollution Estimation.
Xinlei Chen (Carnegie Mellon University) Xiangxiang Xu, Xinyu Liu (Tsinghua University), Shijia Pan (Carnegie Mellon University), Jiayou He (Beijing Experimental High School Attached to Beijing Normal University), Hae Young Noh (Carnegie Mellon University) Lin Zhang (Tsinghua-Berkeley Shenzhen Institute) Pei Zhang (Carnegie Mellon University)
Automated Wireless Localization Data Acquisition and Calibration with 6DOF Image Localization.
Jonathan Fürst (NEC Labs Europe) Kaifei Chen (UC Berkeley) Gürkan Solmaz, Ernö Kovacs (NEC Labs Europe)
Lunch (12:00-13:30)
Session 3: Neural Networks Enhancement (14:00-15:00), Chair: Alberto Gil Ramos (Nokia Bell Labs)
PPG-based Heart Rate Estimation with Time-Frequency Spectra: A Deep Learning Approach.
Attila Reiss, Philip Schmidt (Robert Bosch GmbH) Ina Indlekofer (University Stuttgart) Kristof Van Laerhoven (University Siegen)
Imputation of Missing Data in Time Series for Air Pollutants using Long Short-Term Memory Recurrent Neural Networks.
Hongwu Yuan, Guoming Xu, Zijian Yao (Anhui XinHua University) Ji Jia (Columbia University) Yiwen Zhang (Anhui XinHua University)
Tweet Emoji Prediciton Using Hierarchical Model with Attention.
Chuhan Wu (Tsinghua University) Fangzhao Wu (Microsoft Research Asia) Sixing Wu, Yongfeng Huang (Tsinghua University)
Coffee break (15:00-15:30)
Session 4: Human Information Learning (15:30-16:30), Chair: Carlos Ruiz (CMU)
On Indoor Human Sensing using Commodity Radar.
Mohammed Alloulah, Anton Isopoussu, Fahim Kawsar (Nokia Bell Labs)
Occupant Activity Level Estimation Using Floor Vibration.
Yue Zhang (Tsinghua University) Shijia Pan, Jonathon Fagert, Mostafa Mirshekari, Hae Young Noh, Pei Zhang (Carnegie Mellon University) Lin Zhang (Tsinghua University)
Improving Bag-Of-Words: Capturing Local Information for Motion-Based Activity Recognition
Ming Zeng (Carnegie Mellon University) Helen Qin (The University of North Carolina Chapel Hill) Tong Yu, Chris Lee, Ole J. Mengshoel, John Paul Shen (Carnegie Mellon University)
Summary (16:00-16:30)
THE END
THE VENUE
The CPD 2018 workshop is part of (co-located with) Ubicomp 2018, which will be held at Suntec Convention Center.