서브메뉴
검색
Efficient Learning for Sensor Time Series in Label Scarce Applications
Efficient Learning for Sensor Time Series in Label Scarce Applications
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211152644
- ISBN
- 9798384468615
- DDC
- 610
- 서명/저자
- Efficient Learning for Sensor Time Series in Label Scarce Applications
- 발행사항
- [Sl] : University of California, San Diego, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 132 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-04, Section: B.
- 주기사항
- Advisor: Gupta, Rajesh K.;Shang, Jingbo.
- 학위논문주기
- Thesis (Ph.D.)--University of California, San Diego, 2024.
- 초록/해제
- 요약Sensors in daily life span applications from healthcare and climate modeling to home automation and robotics. These sensors generate abundant time series data, aiding our understanding of various real-life processes. However, much of this data is "unlabeled", that is, without any annotations as to what the data itself refers to in an application. Also, unlike images or text, time series data is challenging for humans to easily interpret and label. Unlike identifying, say objects in a picture, time-series data often requires human experts to analyze using tools and interpret such data before any useful labels can be attached to the data. This is expensive, especially in real-time settings. It is also not scalable due to costs and scarcity of human annotators. Further, labeling data involving human subjects can compromise privacy and security, resulting in a further shortage of labeled data.Self-supervised learning, that is algorithms that learn from unlabeled data, has been used to address the label scarcity problem by automatically generating pseudo-labels from the data itself. However, these algorithms are suboptimal when applied to sensor data because they have not been specifically adapted or customized for the sensory domain. In this dissertation, we will develop methods to adapt traditional self-supervised learning algorithms for sensory domain-specific information to mitigate label scarcity issues. Our approach consists of three steps that can be applied progressively with increased effectiveness. First, we integrate time-interval information of unlabeled data into self-supervised algorithms to build a pre-trained model. Next, we fine-tune the pre-trained model by incorporating application-specific knowledge into a self-supervised algorithm that improves the fine-tuning process. Finally, we propose a sensor context-aware self-supervised algorithm to enhance classical fine-tuning that generalizes to novel classes during testing.We conduct extensive experiments across various sensory data domains, including Motion, Audio, Electroencephalogram (EEG), and Human Activity Recognition (HAR), comparing our methods to leading statistical and deep learning models. By adapting self-supervised algorithms to sensory data with time-awareness, task-specificity, and sensor context-awareness, our methods improve few-shot learning by 10%, fully-supervised learning by 3.6%, and zero-shot learning by 20% compared to the best baselines. Our framework demonstrates state-of-the-art performance across sensing systems of various scales, from small-scale personal healthcare monitoring, human action recognition, and smart home automation to large-scale smart building control, smart city planning, climate modeling, and beyond.
- 일반주제명
- Biomedical engineering
- 일반주제명
- Computer science
- 일반주제명
- Bioinformatics
- 키워드
- Sensors
- 키워드
- Time series data
- 기타저자
- University of California, San Diego Computer Science and Engineering
- 기본자료저록
- Dissertations Abstracts International. 86-04B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017163255
■00520250211152644
■006m o d
■007cr#unu||||||||
■020 ▼a9798384468615
■035 ▼a(MiAaPQ)AAI31485747
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a610
■1001 ▼aChowdhury, Ranak Roy.
■24510▼aEfficient Learning for Sensor Time Series in Label Scarce Applications
■260 ▼a[Sl]▼bUniversity of California, San Diego▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a132 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-04, Section: B.
■500 ▼aAdvisor: Gupta, Rajesh K.;Shang, Jingbo.
■5021 ▼aThesis (Ph.D.)--University of California, San Diego, 2024.
■520 ▼aSensors in daily life span applications from healthcare and climate modeling to home automation and robotics. These sensors generate abundant time series data, aiding our understanding of various real-life processes. However, much of this data is "unlabeled", that is, without any annotations as to what the data itself refers to in an application. Also, unlike images or text, time series data is challenging for humans to easily interpret and label. Unlike identifying, say objects in a picture, time-series data often requires human experts to analyze using tools and interpret such data before any useful labels can be attached to the data. This is expensive, especially in real-time settings. It is also not scalable due to costs and scarcity of human annotators. Further, labeling data involving human subjects can compromise privacy and security, resulting in a further shortage of labeled data.Self-supervised learning, that is algorithms that learn from unlabeled data, has been used to address the label scarcity problem by automatically generating pseudo-labels from the data itself. However, these algorithms are suboptimal when applied to sensor data because they have not been specifically adapted or customized for the sensory domain. In this dissertation, we will develop methods to adapt traditional self-supervised learning algorithms for sensory domain-specific information to mitigate label scarcity issues. Our approach consists of three steps that can be applied progressively with increased effectiveness. First, we integrate time-interval information of unlabeled data into self-supervised algorithms to build a pre-trained model. Next, we fine-tune the pre-trained model by incorporating application-specific knowledge into a self-supervised algorithm that improves the fine-tuning process. Finally, we propose a sensor context-aware self-supervised algorithm to enhance classical fine-tuning that generalizes to novel classes during testing.We conduct extensive experiments across various sensory data domains, including Motion, Audio, Electroencephalogram (EEG), and Human Activity Recognition (HAR), comparing our methods to leading statistical and deep learning models. By adapting self-supervised algorithms to sensory data with time-awareness, task-specificity, and sensor context-awareness, our methods improve few-shot learning by 10%, fully-supervised learning by 3.6%, and zero-shot learning by 20% compared to the best baselines. Our framework demonstrates state-of-the-art performance across sensing systems of various scales, from small-scale personal healthcare monitoring, human action recognition, and smart home automation to large-scale smart building control, smart city planning, climate modeling, and beyond.
■590 ▼aSchool code: 0033.
■650 4▼aBiomedical engineering
■650 4▼aComputer science
■650 4▼aBioinformatics
■653 ▼aSensors
■653 ▼aTime series data
■653 ▼aLabel scarce applications
■653 ▼aSelf-supervised algorithms
■653 ▼aElectroencephalogram
■690 ▼a0800
■690 ▼a0984
■690 ▼a0541
■690 ▼a0715
■71020▼aUniversity of California, San Diego▼bComputer Science and Engineering.
■7730 ▼tDissertations Abstracts International▼g86-04B.
■790 ▼a0033
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17163255▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
