BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CECS - ECPv6.18.4//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://www.cecs.uci.edu
X-WR-CALDESC:Events for CECS
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20230312T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20231105T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20240310T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20241103T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20250309T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20251102T090000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20240524T093000
DTEND;TZID=America/Los_Angeles:20240524T103000
DTSTAMP:20240528T213610Z
CREATED:20240528T213210Z
LAST-MODIFIED:20240528T213610Z
UID:16300-1716543000-1716546600@www.cecs.uci.edu
SUMMARY:Deep Latent Variable Modeling of Physiological Signals
DESCRIPTION:Name: Khuong An Vo \nChair: Prof. Nikil Dutt \nDate: May 24\, 2024 \nTime:  9:30 AM \nLocation: Zoom \nCommittee: Prof. Erik Sudderth\, Prof. Hung Cao\, and Prof. Ramesh Srinivasan \nTitle: Deep Latent Variable Modeling of Physiological Signals \nAbstract: A deep latent variable model is a powerful tool for modeling complex distributions. These models assume that underlying structures\, but unobserved\, are present within the data. In this dissertation\, we explore high-dimensional problems related to physiological monitoring using latent variable models. First\, we present a novel deep state-space model to generate electrical waveforms of the heart using optically obtained signals as inputs. This can bring about clinical diagnoses of heart disease via simple assessment through wearable devices. Second\, we present a brain signal modeling scheme that combines the strengths of probabilistic graphical models and deep adversarial learning. The structured representations can provide interpretability and encode inductive biases to reduce the data complexity of neural oscillations. The efficacy of the learned representations is further studied in epilepsy seizure detection formulated as an unsupervised learning problem. Third\, we propose a framework for the joint modeling of physiological measures and behavior. Existing methods to combine multiple sources of brain data provided are limited. Direct analysis of the relationship between different types of physiological measures usually does not involve behavioral data. Our method can identify the unique and shared contributions of brain regions to behavior and can be used to discover new functions of brain regions. The success of these innovative computational methods would allow the translation of biomarker findings across species and provide insight into neurocognitive analysis in numerous biological studies and clinical diagnoses\, as well as emerging consumer applications. \nShort Bio: Khuong An Vo is a Ph.D. candidate in Computer Science at UC Irvine\, advised by Prof. Ramesh Srinivasan\, Prof. Nikil Dutt\, and Prof. Hung Cao. Prior\, he received a B.Eng. (Honors) degree in CS from HCMC Vietnam National University with research revolved around statistical natural language processing and has been applied in large-scale commercial systems. Now\, Vo’s research focuses on deep latent variable models in biomedical signal processing and neurocognitive modeling. He is also interested in the application of efficient deep learning in healthcare IoT. Vos’ work is supported by National Science Foundation (NSF) and National Institutes of Health (NIH) grants.
URL:https://www.cecs.uci.edu/event/deep-latent-variable-modeling-of-physiological-signals/
LOCATION:Zoom
CATEGORIES:PhD Defenses
ATTACH;FMTTYPE=image/jpeg:https://cecs.uci.edu/files/2024/05/Khuong-An-Vo-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20240528T110000
DTEND;TZID=America/Los_Angeles:20240528T120000
DTSTAMP:20240827T000948Z
CREATED:20240827T000948Z
LAST-MODIFIED:20240827T000948Z
UID:16359-1716894000-1716897600@www.cecs.uci.edu
SUMMARY:Differential Privacy for Non-standard Settings
DESCRIPTION:Name: Joann Chen \nChair: Prof. Zhou Li \nDate: May 28\, 2024 \nTime:  11:00 AM \nLocation: EH 5204 \nCommittee: Prof. Yanning Shen\, Prof. Gene Tsudik \nTitle: Differential Privacy for Non-standard Settings \nAbstract: In today’s technologically driven world\, personal data stands at a crucial juncture\, representing both immense potential and significant peril. Such potential\, when leveraged properly\, can lead to groundbreaking advancements in sectors such as healthcare and finance. Yet\, the misuse of this same data can result in severe privacy breaches. This thesis is dedicated to navigating the complex terrain of personal data privacy\, highlighting Differential Privacy (DP) as a pivotal solution that provides strong\, provable privacy guarantees. \nWith a focus on DP-aware system design\, this thesis delves into the challenges of integrating DP into various systems\, emphasizing the necessity for infrastructures that prioritize privacy preservation. It outlines the progression of DP from theoretical concepts to practical implementations\, detailing both the challenges encountered and the progress made in applying academic DP concepts to real-world solutions. Moreover\, this thesis extends beyond the scope of DP\, exploring the broader realm of privacy-enhancing technologies (PETs). These technologies are crucial not only for meeting legal compliance and ensuring privacy in systems and machine learning but also for paving the way towards a future where data utility and individual privacy coexist in harmony. \nShort Bio: Joann is a fifth-year Ph.D. candidate in the EECS department advised by Prof. Zhou Li. Her research interests center around Differential Privacy (DP)\, privacy-enhancing technologies\, and privacy in machine learning. She has experience in quantifying privacy risks in machine learning and building DP into DNS resolution\, data stream release\, resource allocator\, ad conversion measurements\, and network data synthesis. Her research aims to bridge the gap between DP theory and its practical applications for real-world deployments.
URL:https://www.cecs.uci.edu/event/differential-privacy-for-non-standard-settings/
LOCATION:EH 5204
CATEGORIES:PhD Defenses
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20240531T140000
DTEND;TZID=America/Los_Angeles:20240531T150000
DTSTAMP:20240516T201808Z
CREATED:20240516T201808Z
LAST-MODIFIED:20240516T201808Z
UID:16295-1717164000-1717167600@www.cecs.uci.edu
SUMMARY:Multi-Sensor Data Fusion and Machine Learning for Adaptive Autonomous Systems
DESCRIPTION:Name: Trier Mortlock \nChair: Prof. Mohammad Al Faruque \nDate: May 31\, 2024 \nTime:  2 PM \nLocation: ISEB 1200 \nCommittee: Prof. Faryar Jabbari\, Prof. Pramod Khargonekar \nTitle: Multi-Sensor Data Fusion and Machine Learning for Adaptive Autonomous Systems \nAbstract: Autonomous systems rely on sensors to collect data about their surrounding environments to influence their decision-making algorithms.  Sensor fusion involves combining data from different sensors to reduce sensing uncertainties and better understand the environment. Advances in computing platforms and the growing availability of data have ushered a wave of machined learning-based sensor fusion methods that can extend fusion capabilities beyond the current understanding of systems present in mathematical models. However\, autonomous systems are increasingly tasked with operating in dynamic settings that require high levels of adaptability\, and how best to employ adaptive sensor fusion strategies is a challenging question. This dissertation examines how\, where\, why\, and when to perform multi-sensor data fusion for autonomous systems operating in dynamic environments. A novel sensor fusion architecture that improves autonomous systems’ adaptation capabilities is introduced. The proposed fusion architecture dynamically adapts both how and when fusion is performed by using a machine learning model that learns to identify the optimal sensor fusion strategy for the given sensing context. Use cases across different types of autonomous systems are examined\, highlighting the ability of the proposed fusion approach to generalize across various systems and autonomy tasks. Overall\, the methodologies and ideas presented in this dissertation can be applied broadly across engineering disciplines and can enhance the development of future technologies that require adaptive sensing capabilities. \nShort Bio: Trier Mortlock is a Ph.D. candidate in the Mechanical and Aerospace Engineering program at UC Irvine working under the supervision of Professor Mohammad Al Faruque in the Autonomous and Intelligent Cyber-Physical Systems Lab. Trier obtained his B.S. in Mechanical Engineering from the University of California\, Berkeley and his M.S. in Mechanical and Aerospace Engineering from UC Irvine. During his Ph.D.\, Trier interned at the Frontier Development Lab at the SETI Institute\, NASA and was a Doctoral Research Fellow at the Automotive Research Center\, a U.S. Army Center of Excellence for Modeling and Simulation of Ground Vehicle Systems.
URL:https://www.cecs.uci.edu/event/multi-sensor-data-fusion-and-machine-learning-for-adaptive-autonomous-systems/
LOCATION:ISEB 1200\, Interdisciplinary Science and Engineering Building University of California\, Irve\, Irvine\, CA\, 92697\, United States
CATEGORIES:PhD Defenses
END:VEVENT
END:VCALENDAR