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:20240402T120000
DTEND;TZID=America/Los_Angeles:20240402T130000
DTSTAMP:20240502T215645Z
CREATED:20240502T215253Z
LAST-MODIFIED:20240502T215645Z
UID:16268-1712059200-1712062800@www.cecs.uci.edu
SUMMARY:Fairness-Aware learning over Graphs
DESCRIPTION:Title: Fairness-Aware learning over Graphs \nSpeaker: Oyku Deniz Kose \nDate and Time: Tuesday\, April 2nd\, 12:00pm \nLocation: EH 2430 \nAbstract: \nGraphs are powerful mathematical tools that can represent complex real-world systems\, such as financial markets and social networks\, and the relations within them. Due to this\, machine learning (ML) over graphs has attracted significant attention recently\, and showed promising success in various applications. Despite the success\, the large-scale deployment of graph-based ML algorithms in real-world systems relies heavily on how socially responsible they are. While graph-based ML models nicely integrate the nodal data with the connectivity\, they also inherit potential unfairness. Using such ML models may therefore result in inevitable unfair results in various decision- and policy-making in the related applications. While fairness and explainability have attracted increasing attention in responsible ML\, they are mostly under-explored in the graph domain. Motivated by this\, in this talk\, we first focus on the analysis of sources of bias in graph-based learning and then present fairness-aware learning algorithms that combat the identified bias sources. In addition\, we will talk about certain real-word challenges that need to be considered in addition to fairness\, e.g.\, efficiency\, convergences. \nBiography: \nOyku Deniz Kose received the B.S. and M.S. degrees in electrical and electronics engineering from Bogazici University\, Istanbul\, Turkey\, in 2017 and 2020\, respectively. She is currently working towards a Ph.D. degree with the Department of Electrical Engineering and Computer Science\, University of California Irvine\, Irvine\, CA\, USA. She received the Samueli Endowed Fellowship in 2021 and was a research intern at Google in 2023. Her research mainly focuses on trustworthy machine learning with a specific interest in graph-based learning.
URL:https://www.cecs.uci.edu/event/fairness-aware-learning-over-graphs/
LOCATION:EH 2430\, Engineering Hall\, University of California\, Irvine\, Irvine\, CA\, 92697\, United States
CATEGORIES:Student Seminars
ATTACH;FMTTYPE=image/jpeg:https://cecs.uci.edu/files/2024/05/Kose-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20240123T120000
DTEND;TZID=America/Los_Angeles:20240123T130000
DTSTAMP:20240502T211540Z
CREATED:20240502T211540Z
LAST-MODIFIED:20240502T211540Z
UID:16263-1706011200-1706014800@www.cecs.uci.edu
SUMMARY:Provably-Safe Offloading of Neural Network Controllers for Energy-Efficiency in Autonomous Driving Systems
DESCRIPTION:Title: Provably-Safe Offloading of Neural Network Controllers for Energy-Efficiency in Autonomous Driving Systems \nSpeaker: Mohanad Odema \nDate and Time: Tuesday\, January 23rd\, 12:00pm \nLocation: EH 2430 \nAbstract: \nTo mitigate the high energy demand of Neural Network (NN) based Autonomous Driving Systems (ADSs)\, we consider the problem of offloading NN controllers from the ADS to nearby edge-computing infrastructure\, but in such a way that formal vehicle safety properties are guaranteed. In particular\, we propose the EnergyShield framework\, which repurposes a controller “shield” as a low-power runtime safety monitor for the ADS vehicle. Specifically\, the shield in EnergyShield provides not only safety interventions but also a formal\, state-based quantification of the tolerable edge response time before vehicle safety is compromised. Using EnergyShield\, an ADS can then save energy by wirelessly offloading NN computations to edge computers\, while still maintaining a formal guarantee of safety until it receives a response (on-vehicle hardware provides a just-in-time fail safe). To validate the benefits of EnergyShield\, we implemented and tested it in the Carla simulation environment. Our results show that EnergyShield maintains safe vehicle operation while providing significant energy savings compared to on-vehicle NN evaluation: from 24% to 54% less energy across a range of wireless conditions and edge delays. \nBiography: \nOdema received a B.Sc degree in Electronics and Communications in 2014 and a M.Sc degree in Computer Engineering in 2018 from Ain Shams University\, Egypt. He is currently pursuing a Ph.D. degree in Computer Engineering at UCI where his current research interests are focused on HW/SW co-design methodologies for enhancing the performance efficiency of key AI workloads as from vision and NLP applications in emerging domains such as Edge-AI\, autonomous systems\, and tinyML.
URL:https://www.cecs.uci.edu/event/provably-safe-offloading-of-neural-network-controllers-for-energy-efficiency-in-autonomous-driving-systems/
LOCATION:EH 2430\, Engineering Hall\, University of California\, Irvine\, Irvine\, CA\, 92697\, United States
CATEGORIES:Student Seminars
ATTACH;FMTTYPE=image/jpeg:https://cecs.uci.edu/files/2024/05/odema.jpg
END:VEVENT
END:VCALENDAR