Workshop on Challenges in Artificial Intelligence and Machine Learning for Internet of Things (AIChallengeIoT)

The 4th International Workshop on Challenges in Artificial Intelligence and Machine Learning for Internet of Things (AIChallengeIoT 2022) will be held in conjunction with ACM SenSys 2022.

Artificial intelligence (AI) and machine learning (ML) are key enabling technologies for many Internet of Things (IoT) applications. However, the collection and processing of data for AI and ML is very challenging in the IoT domain. For example, there are usually a large number of low-powered sensors deployed in large geographical areas with possibly intermittent network connectivity. The sensors and their collected data may be owned by different users or organizations, which can bring further obstacles to data collection due to privacy concerns and noisy labels provided by different users. The successful application of AI/ML approaches in such scenarios with noisy and decentralized data is difficult. In addition, the amount of collected data that can be used for training AI/ML models is usually proportional to the number of users in the system, but the system may not be able to attract many users without a well-trained AI/ML model, and it is challenging to solve this dilemma.

This workshop focuses on how to address the above and other unique challenges of applying AI/ML in IoT systems.


Workshop Program (Sunday, November 6, 2022)


8:00 - 9:00: Arrival (Venue: Room Maverick-A, Hilton Boston Back Bay)


9:00 - 9:10: Welcome and Opening Remarks (Luis Garcia)


Technical Session 1
Session chair: Luis Garcia

9:10 - 9:35: (Best Paper) Ultra-low Power DNN Accelerators for IoT: Resource Characterisation of the MAX78000
Arthur Moss (Nokia Bell Labs and Newcastle University); Hyunjong Lee (Nokia Bell Labs and KAIST); Lei Xun (Nokia Bell Labs and University of Southampton); Chulhong Min, Fahim Kawsar, Alessandro Montanari (Nokia Bell Labs)

9:35 - 10:00: ElasticAI-Creator: Optimizing Neural Networks for Time-Series-Analysis for On-Device Machine Learning in IoT
Chao Qian, Lukas Einhaus, Gregor Schiele (University of Duisburg-Essen)


10:00 - 10:20: Networking & Morning Break

Technical Session 2
Session chair: Luis Garcia

10:20 - 10:45: Smart Objects: Impact localization powered by TinyML
Ioannis Katsidimas (Department of Computer Engineering and Informatics, University of Patras, Greece); Thanasis Kotzakolios (Department of Mechanical Engineering and Aeronautics, University of Patras, Greece); Sotiris Nikoletseas (Department of Computer Engineering and Informatics, University of Patras, Greece and Computer Technology Institute and Press “Diophantus”, Greece); Stefanos H. Panagiotou, Constantinos Tsakonas (Department of Computer Engineering and Informatics, University of Patras, Greece)

10:45 - 11:10: The Impact of Cascaded Optimizations in CNN models and End-device Deployment
Hanan Hussain, PS Tamizharasan (BITS Pilani, Dubai Campus, UAE)


11:10 - 11:25: Networking & Morning Break


11:25 - 11:30: Best Paper Award Announcement

Best Paper #1: Ultra-low Power DNN Accelerators for IoT: Resource Characterisation of the MAX78000, Arthur Moss et al.

Best Paper #2: Increasing the Intelligence of Low Power Sensors with Autonomous Agents, Jannik William et al.

11:30 - 12:20: Keynote: Creating sustainable, Data-driven Buildings
Keynote speech by Bharathan Balaji, Amazon AI Lab


12:20 - 12:30: Questions & Networking



12:30 - 14:00: Lunch Break


Technical Session 3
Session chair: Shabih Hasan (Delos)

14:00 - 14:25: Evaluation of ASR for Conversational Speech in Voice Assistants: A Linguistic Perspective
Hannaneh B. Pasandi (Virginia Commonwealth University); Haniyeh B. Pasandi (University of Maryland Baltimore County)

14:25 - 14:50: CIPhy: Causal Intervention with Physical Confounder from IoT Sensor Data for Robust Occupant Information Inference
Zhizhang Hu (University of California, Merced); Tong Yu (Carnegie Mellon University); Ruiyi Zhang (Duke University); Shijia Pan (University of California, Merced)

14:50 - 15:15: CycleGAN based Unsupervised Domain Adaptation for Machine Fault Diagnosis
Naibedya Pattnaik (Embedded Devices and Intelligent Systems, TCS Research, Bangalore, India); Uday Sai Vemula (Indian Institute of Technology Madras, Chennai, India); Kriti Kumar, A. Anil Kumar (Embedded Devices and Intelligent Systems, TCS Research, Bangalore, India); Angshul Majumdar (Indraprastha Institute of Information Technology Delhi, New Delhi, India); M.Girish Chandra, Arpan Pal (Embedded Devices and Intelligent Systems, TCS Research, Bangalore, India)


15:15 - 15:30: Afternoon Break


Technical Session 4
Session chair: Dezhi Hong

15:30 - 15:55: Federated Learning Biases in Heterogeneous Edge-Devices - A Case-study
Khotso Selialia, Yasra Chandio, Fatima M. Anwar (University of Massachusetts Amherst)

15:55 - 16:20: Security Analysis of SplitFed Learning
Momin Ahmad Khan, Virat Shejwalkar (University of Massachusetts Amherst); Amir Houmansadr (UMass Amherst); Fatima M. Anwar (University of Massachusetts Amherst)

16:20 - 16:45: (Best Paper) Increasing the Intelligence of Low Power Sensors with Autonomous Agents
Jannik William (ETH Zürich); Matuzalem Muller dos Santos, Jomi Fred Hübner, Maiquel Brito (Universidade Federal de Santa Catarina); Danai Vachtsevanou, Andres Gomez (University of St.Gallen)

16:45 - 17:00: Closing Remarks (Luis Garcia)

Workshop History

Organizing Committee

Program Chairs
Steering Committee
Program Committee