AIoT Workshop Program @ KDD’21

August 15th, 2021 (8am-12pm, SG Time)
Location: Virtual
8:00 - 8:10am


8:10 - 8:40am

Keynote: Advances in Distributed Video Analytics

Vijaykrishnan Narayanan, Penn State University

Professor Vijaykrishnan Narayanan is a Robert Noll Chair Professor of Computer Science & Engineering and Electrical Engineering at the Pennsylvania State University. Vijay received his Bachelors in Computer Science & Engineering from University of Madras, India in 1993 and his Ph.D. in Computer Science & Engineering from the University of South Florida, USA, in 1998. He is a co-director of the Microsystems Design Lab at Penn State. He is a Fellow of IEEE, ACM and National Academy of Inventors.

Abstract: This talk will introduce innovations in multi-view video analytic systems. First, I will discuss our explorations on design of communication-aware distributed inference techniques. Next, I will discuss our processing-in-memory architectural techniques that transform on-chip SRAM caches for accelerating machine learning algorithms. Finally, I will conclude with interesting open challenges.

Presentation deck [PDF]

Sasha Mitarai is the Section Head of Modeling & Optimization in the Process Technology Department at ExxonMobil’s R&D center in Clinton, NJ. She has 10+ years of experience which includes managing research and engineering teams to develop and deliver state of art high performance computing and data science solutions. Sasha holds a Master’s in Information Systems Operations Management and a Bachelor’s in Business from University of Florida.

Abstract: As Data Science advances and increasingly large amounts of data become readily available via various Industrial Internet of Things (IIoT) projects, the energy industry is going through a digital transformation to unlock opportunities. In this talk, I will provide an overview and share two case studies of Digital Transformation enabled by Data Science R&D (1) Manufacturing Data Lake– efficient access to global manufacturing data (2) DEEVA: a deep learning and IoT based computer vision system to address safety and security of production sites.

9:05 - 9:50am

Technical Paper Session

Sequential Self-Adaptation to Evolving Users in Real-Time for Wearable-Based Human Activity Recognition
Hangwei Qian, Chunyan Miao (Nanyang Technological University, Singapore)
Early Mobility Recognition for Intensive Care Unit Patients Using Accelerometers
Rex Liu, Sarina A Fazio, Huanle Zhang, Albara Ah Ramli, Xin Liu, Jason Yeates Adams (University of California Davis, USA)
Assessing Machine Learning Approaches to Address IoT Sensor Drift
Haining Zheng, Antonio R. Paiva (ExxonMobil Research and Engineering Company, USA)

Kenneth Tran is the founder and CEO of Koidra Inc. Before Koidra, Kenneth was a Principal Applied Scientist in the Machine Learning Group, Microsoft Research (MSR). While at MSR, he led the Sonoma team, winning the first autonomous greenhouse challenge (2018), becoming the only AI team that outperformed expert Dutch growers. He recently led the Koala team comprised of Koidra and Cornell University researchers to win the first phase of the 3rd autonomous greenhouse challenge in 2021.Koala outperformed 46 teams from 24 countries in both the virtual net profit and overall ranking.

Ken's research expertise and experience include Reinforcement Learning, Deep Learning, Optimization, and Distributed Computing. At Microsoft, he led the research and development for strategic AI projects such as Deep Reinforcement Learning for real-world control problems, Computer Vision API for Cognitive Services, and Autonomous Greenhouses.

Kenneth received his Ph.D. in Computational & Applied Mathematics from The University of Texas at Austin in 2012.

Abstract: AIoT holds a great promise to disrupt greenhouse agriculture as well as operational control in manufacturing in general. Optimizing the conditions for plant growth is one of the most challenging operational control problems because it involves biological processes which we humans still don’t know much about. For this talk, I'll present: 1) The case of autonomous greenhouses using AIoT, in which I have won the international autonomous greenhouse challenges twice (in 2018 & 2021) and also outperformed the expert growers; 2) How our team is tackling the challenges in operational control in greenhouses and propose a unified framework for solving these issues with AIoT.

Presentation deck [PDF]

Dr. Jie Liu Dean of AI Research in Harbin Institute of Technology. Prior to joining HIT in March 2019, he was a Partner Research Manager at Microsoft AI& Research, leading a research team in the Ambient Intelligence area of Microsoft Business AI (BizAI). His research interests root in understanding and managing the physical properties of computing. Examples include timing, location, energy, and the awareness of and impact on the physical world. I worked on actor-oriented modeling, simulation, & programming models, low power and networked sensing, system resource management & control, mobile context extraction & management, and novel mobile & cloud applications. He received 5 Best Paper Awards from conferences like SenSys, MobiSys, RTSS, and RTAS. Here is my full publication list. Dr. Liu is an IEEE Fellow, an ACM Distinguished Scientist, and an ACM Distinguished Speaker. He serves as the Steering Committee Chair for Cyber-Physical Systems Week (CPS Week), Steering Committee Chair for ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), and a Steering Committee member of ACM International Conference on Embedded Networked Sensing System (SenSys). He was an Associate Editor for IEEE Trans. on Mobile Computing and ACM Trans. on Sensor Networks.

Presentation deck [PDF]

11:00am - 12:00pm
Moderator: Yuanchao Shu, Microsoft

Microsoft sponsored and co-organized Indoor Location Competition 2.0 in 2021. 1446 contestants from more than 60 countries making up 1170 teams participated in this unique global event. In this competition, a first-of-its-kind large-scale indoor location benchmark dataset was released. In this panel, competition winners and organizers will share their experience and findings, and join with location-based service practitioners from industry and academia to discuss opportunities and challenges in the broader location sensing area.

More details on the competition including winning solutions can be found on our Kaggle competition site.”

Yuanchao Shu is currently a Senior Researcher with Microsoft. His research interests lie broadly in mobile, sensing and networked systems, with a main focus on applications including large-scale video analytics and location-based systems. His previous research results have led to over 40 publications at top-tier peer-reviewed scientific conferences and journals including MobiCom, MobiSys, NSDI, SenSys, USENIX Security, JSAC, TMC, TPDS etc. Dr. Shu currently serves on the editorial board of ACM Transactions on Sensor Networks, and was a member of the organizing committee and TPC of conferences including MobiCom, SenSys, SEC, Globecom, ICC etc. He won five Best Paper/Demo (Runner-Up) Awards, and was the recipient of ACM China Doctoral Dissertation Award and IBM PhD Fellowship. Dr. Shu received his Ph.D. from Zhejiang University, and was also a joint Ph.D. student in the EECS Department at the University of Michigan, Ann Arbor.


Dmitry Gordeev and the team he belongs won the first place in Indoor Location Competition 2.0. Dmitry is a Senior Data Scientist at and a Kaggle Grandmaster, ranked 5th worldwide. He graduated from Moscow State University with honors as a specialist in data mining. He has 10 years of professional experience in credit risk management and more than 3 years of experience competing on Kaggle. Dmitry won competitions on laboratory earthquake forecasting, texts classification, sports analytics and earned gold medals in tabular, NLP, computer vision, signal classification and reinforcement learning competitions.

Reza Vaghefi and the team he belongs won the second place in Indoor Location Competition 2.0. Reza received the M.S. degree in communication engineering from Chalmers University of Technology, Gothenburg, Sweden, in 2011 and the Ph.D. degree in electrical engineering from Virginia Tech, Blacksburg, VA, USA, in 2014. He has 3 patents and more than 30 publications in the field of localization and tracking. His research interests include machine learning, cooperative and noncooperative localization, wireless communications and signal processing, estimation theory, and convex optimization. Reza serves as an Associate Editor for EURASIP Journal on Advances in Signal Processing, and Wireless Communications and Mobile Computing. He is currently a research scientist in Qualcomm working on the next generation Wi-Fi chipset.

Yves Miehe won the third place in Indoor Location Competition 2.0. Yves is a research and development engineer, focused on simulation. He has worked for several years in an insurance company as a flood model engineer, and for several car and engine companies. He is now working at MCE-5 development, a small company based in Lyon, France, where he simulates the combustion of engines, and the behavior of several components for the engine. He invented several patents to improve the fuel consumption of combustion engines, one of which is currently being developed as a prototype by the company. Yves had his engineer degree in year 2001 at Ecole Centrale Paris. He got interested by statistics and modeling in his various professional experiences, and discovered Kaggle in 2015 for a driver analysis competition. Despite not being a data scientist, his simulation experiences enable him to find solutions on Kaggle data science competitions. After 12 entertaining competitions, including 6 gold medals, he got the title of “competition grand master” at the end of the Indoor location competition.

Qiang is the founder and the chief scientist of XYZ10, which is a startup company offering a reliable indoor-outdoor positioning service. Qiang received his Ph.D from the University of Michigan in 2013. Before founding XYZ10, Qiang has been working at NEC Labs Amercia, Boeing Labs, Microsoft Research, AT&T Labs Research. Along this career path, Qiang has been collected diverse hands-on experience regarding systems, networking, AIoT, deep learning, etc.

Manikanta Kotaru is a Senior Researcher with Azure for Operators Office of CTO. His expertise is in wireless networks, radio frequency machine vision and computational sensing. He graduated from Stanford University and was advised by Prof. Sachin Katti. During his PhD, he developed world’s first centimeter-level accurate wireless positioning using commodity WiFi devices. Before joining Stanford, he graduated from IIT Bombay with Bachelors in Electrical Engineering and minor in Computer Science. He won Silver medal at International Chemistry Olympiad and is a Thomas and Sarah Kailath Stanford Graduate Fellow.

Need to find us?