ABSTRACT
The number of Internet-of-Things (IoT) and edge devices has exploded in recent years. Coupled with recent advances in learning methodologies, these can make the vision of smart building a reality and transform how people interact with their environment. Deep learning-driven systems have been already successful in many IoT application areas of pervasive sensing and ubiquitous computing including human activity monitoring and occupancy detection. However, the robustness of such systems at scale is dependent on the availability of a large amount of data labeled by a human expert. Acquiring human annotation involves high effort, high cost, and is often error-prone. Therefore, learning to execute tasks in a machine learning system without manual inspection is of prime interest as it can open up opportunities for continual learning in the long-term deployment of IoT systems. In this work, we propose deep neural network-based approaches for segmentation and classification tasks in sensor streams with no or limited labels. We focus on reducing human labeling efforts in machine learning systems by 1) leveraging temporal features learned through deep neural networks for downstream tasks and 2) learning robust features in dynamic environments by utilizing multiple sensing sources and semi-supervised learning (e.g. small amount of labeled data).
- Wei-Cheng Chang, Chun-Liang Li, Yiming Yang, and Barnabás Póczos. 2019. Kernel Change-point Detection with Auxiliary Deep Generative Models. In International Conference on Learning Representations.Google Scholar
- Raia Hadsell, Dushyant Rao, Andrei A Rusu, and Razvan Pascanu. 2020. Embracing Change: Continual Learning in Deep Neural Networks. Trends in Cognitive Sciences (2020).Google Scholar
- Gierad Laput, Yang Zhang, and Chris Harrison. 2017. Synthetic Sensors: Towards General-Purpose Sensing. In 2017 CHI Conference on Human Factors in Computing Systems (Denver, Colorado, USA) (CHI '17). Association for Computing Machinery, New York, NY, USA, 3986--3999. https://doi.org/10.1145/3025453.3025773Google ScholarDigital Library
- Shuang Li, Yao Xie, Hanjun Dai, and Le Song. 2015. M-Statistic for Kernel Change-Point Detection. In 28th International Conference on Neural Information Processing Systems (NIPS'15, Vol. 2). 3366--3374.Google Scholar
- Hongfei Xue, Wenjun Jiang, Chenglin Miao, Fenglong Ma, Shiyang Wang, Ye Yuan, Shuochao Yao, Aidong Zhang, and Lu Su. 2020. DeepMV: Multi-View Deep Learning for Device-Free Human Activity Recognition. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 4, 1, Article 34 (March 2020), 26 pages.Google ScholarDigital Library
- Yang Zhang, Yasha Iravantchi, Haojian Jin, Swarun Kumar, and Chris Harrison. 2019. Sozu: Self-Powered Radio Tags for Building-Scale Activity Sensing. In 32nd Annual ACM Symposium on User Interface Software and Technology (New Orleans, LA, USA) (UIST '19). 973--985.Google ScholarDigital Library
- Towards Reducing Labeling Efforts in IoT-based Machine Learning Systems: PhD Forum Abstract
Recommendations
A top-down survey on securing IoT with machine learning: goals, recent advances and challenges
The Internet of Things (IoT) has seen it all from being just another innovation to a leading technology; it is now a binding force that interconnects various aspects of our lives. The IoT's tremendous growth is driven by emerging applications and evolving ...
Machine Learning: The State of the Art
The two fundamental problems in machine learning (ML) are statistical analysis and algorithm design. The former tells us the principles of the mathematical models that we establish from the observation data. The latter defines the conditions on which ...
Comments