BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Information Systems Group - ECPv6.4.0.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://isg.ics.uci.edu
X-WR-CALDESC:Events for Information Systems Group
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:20240310T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20241103T090000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20240405T130000
DTEND;TZID=America/Los_Angeles:20240405T140000
DTSTAMP:20260426T052859
CREATED:20240401T183244Z
LAST-MODIFIED:20240401T183244Z
UID:1696-1712322000-1712325600@isg.ics.uci.edu
SUMMARY:Lukasz Golab (University of Waterloo): Understanding models and the data they learn from
DESCRIPTION:Lukasz Golab (U. Waterloo) \nUnderstanding models and the data they learn from \nAbstract: The modern world is powered by data. However\, as the capabilities of data-intensive systems grow\, so does their complexity\, making them hard to understand and troubleshoot. I will discuss my lab’s efforts towards understanding models and the data they learn from\, including local and global model explanations as well as model diagnostics for fairness and bias avoidance. \n  \nBio: Lukasz Golab is a Professor and Canada Research Chair at the University of Waterloo. From 2006 to 2011\, he was a Senior Member of Research Staff at AT&T Labs. He obtained a BSc in Computer Science from the University of Toronto (with High Distinction) and a PhD in Computer Science from the University of Waterloo (with Alumni Gold Medal). His long-term research agenda of Data for Good calls for building data-intensive systems with societal impact. His recent projects focus on systems for managing high-speed data events such as data stream engines and blockchains\, understanding complex models and the data they learn from\, and applications including online safety\, education\, and sustainability.
URL:https://isg.ics.uci.edu/event/lukasz-golab-university-of-waterloo-understanding-models-and-the-data-they-learn-from/
LOCATION:DBH 4011
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20240412T130000
DTEND;TZID=America/Los_Angeles:20240412T140000
DTSTAMP:20260426T052859
CREATED:20240409T184916Z
LAST-MODIFIED:20240409T185018Z
UID:1703-1712926800-1712930400@isg.ics.uci.edu
SUMMARY:Juncheng Fang: ImmortalChopper: Real-Time and Resilient Distributed Transactions in the Edge-Cloud
DESCRIPTION:Abstract:\n\nEmerging applications in the areas of real-time Internet of Things (IoT) and edge technologies (such as wearables\, and mobile headsets) require fast processing and response times. This motivates the utilization of edge nodes for both processing and storage of data. In settings with a vast number of edge nodes—such as the case of smart cities and spaces—the state of the data is distributed across a large number of edge nodes. This makes it expensive to perform distributed transactions as these transactions would span edge nodes that are connected via less reliable and relatively slow network infrastructure. This makes it prohibitive to use existing protocols like 2PC that require rounds of communication across participants. \nIn this paper\, we propose ImmortalChopper\, a distributed transaction processing protocol that is designed for the edge-cloud environment. The goal of ImmortalChopper is to allow fast commitment of transactions on the edge without having to wait for distributed coordination. To achieve this\, we build on the literature of Transaction Chopping. Transaction Chopping allows breaking a transaction into smaller hops. If the first hop commits\, then\, the rest of the transaction is guaranteed to commit. We utilize this feature to allow a transaction to commit from the closest edge node without having to wait for the rest of the processing of the other participating edge nodes. However\, the direct use of Transaction Chopping is not suitable for the edge-cloud. This is because of the sporadic availability of edge nodes that leads to either blocking behavior during failures or the necessity to replicate each step which defies the purpose of using Chopping in our case. The innovation in ImmortalChopper is the introduction of the concept of ChopperGraph which utilizes lazy replication between edge and cloud nodes. This enables resilience to failures without the added synchronous overhead. \nBio:\nJuncheng Fang is a 3rd-year Ph.D. student in the Computer Science Department at UC Irvine\, supervised by Prof. Faisal Nawab. His current research focuses on blockchain\, distributed systems\, and edge cloud.
URL:https://isg.ics.uci.edu/event/juncheng-fang-immortalchopper-real-time-and-resilient-distributed-transactions-in-the-edge-cloud/
LOCATION:DBH 4011
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20240419T130000
DTEND;TZID=America/Los_Angeles:20240419T140000
DTSTAMP:20260426T052859
CREATED:20240407T214847Z
LAST-MODIFIED:20240407T214847Z
UID:1698-1713531600-1713535200@isg.ics.uci.edu
SUMMARY:Mohammed Al-Kateb (Amazon Redshift): The Evolution of Amazon Redshift
DESCRIPTION:Abstract:\nIn this talk\, we will discuss the evolution of Amazon Redshift over the past 10 years. We’ll discuss the Amazon Redshift architecture. We’ll dive deep in the lifecycle of executing a query in Amazon Redshift. And we’ll examine how Amazon Redshift continues to maintain a leading price/performance in the market. \nBio:\n Mohammed Alkateb leads the Query Optimizer team of Redshift – The Amazon AWS Distributed Cloud Data Warehouse that tens of thousands of customers rely on to gain the insight they need from their most critical data. Prior to joining Amazon\, Mohammed spent over a decade with the Teradata Optimizer team as an individual contributor and engineering manager. Mohammed is also an adjunct professor at Worcester Polytechnic Institute (WPI) and at California State University\, Northridge (CSUN). Mohammed has 16 U.S. patents. And he has publications in research and industrial tracks of premier database conferences including EDBT\, ICDE\, SIGMOD and VLDB. Mohammed holds a Ph.D. degree in Computer Science from The University of Vermont\, and M.Sc. & B.Sc. degrees in Information Systems from Cairo University.
URL:https://isg.ics.uci.edu/event/mohammed-al-kateb-amazon-redshift-the-evolution-of-amazon-redshift/
LOCATION:DBH 4011
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20240426T130000
DTEND;TZID=America/Los_Angeles:20240426T140000
DTSTAMP:20260426T052859
CREATED:20240424T002712Z
LAST-MODIFIED:20240424T002712Z
UID:1757-1714136400-1714140000@isg.ics.uci.edu
SUMMARY:Xinyuan Lin: Data Science Tasks Implemented with Scripts versus GUI-Based Workflows: The Good\, the Bad\, and the Ugly.
DESCRIPTION:Abstract: As leveraging large-scale data analytics becomes the norm for many applications\, platforms for developing these capabilities have become increasingly important. This work compares the benefits and drawbacks of implementing two commonly used data science platform paradigms: code-based scripts and GUI-based workflows. We implement tasks in both paradigms that provide examples of phases in the typical life cycle of a data science project\, including data wrangling\, machine learning (ML) model training\, and inference. In this talk\, we will examine the relative performance of the implementations under each paradigm in various experimental settings. We will discuss the benefits and drawbacks of each platform implementation and provide a foundation for future work in comparing data science platform paradigms. \nBio: Xinyuan Lin is a third-year Ph.D student in the Computer Science Department at UC Irvine. His research interests include data processing systems and big data analytics.
URL:https://isg.ics.uci.edu/event/xinyuan-lin-data-science-tasks-implemented-with-scripts-versus-gui-based-workflows-the-good-the-bad-and-the-ugly/
LOCATION:DBH 4011
END:VEVENT
END:VCALENDAR