BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Events - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Events
X-ORIGINAL-URL:https://events.ucsc.edu
X-WR-CALDESC:Events for Events
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:20250309T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20251102T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20260308T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20261101T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20270314T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20271107T090000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260427T110000
DTEND;TZID=America/Los_Angeles:20260427T130000
DTSTAMP:20260428T120807
CREATED:20260420T225301Z
LAST-MODIFIED:20260423T210320Z
UID:10012119-1777287600-1777294800@events.ucsc.edu
SUMMARY:Quality First Coding Contest
DESCRIPTION:This is a programming contest\, but with a twist! Instead of scoring you based on your speed and solution accuracy\, we score you based on your programming quality and solution accuracy. This means that instead of looking at how fast you can program a solution\, we look at your number of compiles/runs instead.* The contestant that uses the least number of compiles/runs to produce passing code is the winner. Ties are broken by time. \nFood will be provided. QFCC 20260427 – Poster
URL:https://events.ucsc.edu/event/quality-first-coding-contest/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/04/Quality-First-Coding-Contest.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260423T180000
DTEND;TZID=America/Los_Angeles:20260423T193000
DTSTAMP:20260428T120807
CREATED:20260402T213440Z
LAST-MODIFIED:20260402T222539Z
UID:10012030-1776967200-1776972600@events.ucsc.edu
SUMMARY:Climate Week Tech Connect: Energy Solutions
DESCRIPTION:Join Baskin Engineering to explore the frontier of power engineering\, where the rapid rise of electrification and digital infrastructure is creating an unprecedented demand for next-generation talent and a critical opportunity for sustainability.  \nThis networking event bridges the gap between the classroom and the field\, offering students and faculty a front-row seat to the trends and high-impact career opportunities shaping our energy future. The event is part of Baskin Engineering Climate Week\, focused on raising awareness of climate issues and sustainability research and teaching. \nWhere: BE Courtyard\nWhen: Thursday\, April 23\, 6:00-7:30 p.m. \nWe hope to see you there!
URL:https://events.ucsc.edu/event/climate-week-tech-connect-energy-solutions/
LOCATION:Jack Baskin Engineering\, Baskin Engineering 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Meetings & Conferences
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/01/BElogoWHITE.png
GEO:37.000369;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Jack Baskin Engineering Baskin Engineering 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Baskin Engineering 1156 High Street:geo:-122.0632371,37.000369
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260423T170000
DTEND;TZID=America/Los_Angeles:20260423T181500
DTSTAMP:20260428T120807
CREATED:20260402T211703Z
LAST-MODIFIED:20260402T212222Z
UID:10011935-1776963600-1776968100@events.ucsc.edu
SUMMARY:Careers in Climate Tech & Sustainability
DESCRIPTION:Ready to explore career pathways that matter? \nAttend our very special Careers in Climate Tech & Sustainability Panel—celebrating Baskin Engineering Climate Week—for an inside look at careers that will help build a sustainable future. Panelists representing different roles and organizations will share their career journeys and offer practical insights into working in climate tech. There will also be a catered networking reception that follows—don’t miss it! \nGet informed\, inspired\, and discover your path to a career in sustainability! \nThis event is part of Baskin Engineering’s Climate Tech Day featuring a community fair where students\, faculty\, climate and sustainability tech companies\, and community organizations will showcase their works through demonstrations\, poster presentations\, tabling\, and more.  \nWhere: E2-180\nWhen: Thursday\, April 23\, 5:00-6:15 p.m. \nRegister via Handshake. \nIf you have disability-related needs\, please contact the Career Success office at csuccess@ucsc.edu or (831) 459-4420 as soon as possible. \nYOU BELONG HERE\nPrograms and services are open to all\, consistent with state and federal law\, as well as the University of California’s nondiscrimination policies. Every initiative—whether a student service\, faculty program\, or community event—is designed to be accessible\, inclusive\, and respectful of all identities. To learn more\, please visit UC Nondiscrimination Statement or Nondiscrimination Policy for UC Publications.
URL:https://events.ucsc.edu/event/careers-in-climate-tech-sustainability/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/04/Javier-drone.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260423T140000
DTEND;TZID=America/Los_Angeles:20260423T170000
DTSTAMP:20260428T120807
CREATED:20260403T215527Z
LAST-MODIFIED:20260403T220700Z
UID:10012043-1776952800-1776963600@events.ucsc.edu
SUMMARY:Climate Tech & Sustainability Showcase
DESCRIPTION:Join us for a Climate Tech & Sustainability Showcase\, where students\, faculty\, climate and sustainability-focused companies\, founders\, and community organizations come together to share their work and ideas. The event is part of Baskin Engineering Climate Week\, focused on raising awareness of climate issues and sustainability research and teaching. \nExplore a range of interactive demos\, poster presentations\, and tabling displays highlighting innovative research\, emerging technologies\, and real-world solutions to climate challenges. Baskin Engineering student organizations will also be on hand to share their climate friendly projects! \nCome network\, promote your organization\, and meet up-and-coming talent alongside other passionate\, like-minded members of the climate and sustainability community. \nWhere: BE Courtyard\nWhen: 2:00-5:00 p.m.
URL:https://events.ucsc.edu/event/climate-tech-sustainability-showcase/
LOCATION:Jack Baskin Engineering\, Baskin Engineering 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/04/4-23-25-climate-week-cl-002-scaled.jpg
GEO:37.000369;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Jack Baskin Engineering Baskin Engineering 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Baskin Engineering 1156 High Street:geo:-122.0632371,37.000369
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260422T110000
DTEND;TZID=America/Los_Angeles:20260422T121500
DTSTAMP:20260428T120807
CREATED:20260331T171056Z
LAST-MODIFIED:20260401T165930Z
UID:10011819-1776855600-1776860100@events.ucsc.edu
SUMMARY:CSE Colloquium - Robust Machine Learning for Biomedical Data: Efficiency\, Reliability\, and Generalizability
DESCRIPTION:Presenter\nChenyu You\, Stony Brook University \nAbstract\nIn the rapidly growing area of machine learning\, there is profound promise in crafting intelligent\, data-driven methods for diverse real-world applications. Yet\, in safety-critical domains like healthcare\, some fundamental challenges remain: (1) The insufficiency of raw biomedical data emphasizes the need for data-efficient and robust learning approaches. (2) The imperative of safety and stability necessitates a cohesive framework that unifies learning with theoretical guarantees. (3) The inherent heterogeneity and distribution shifts in real-world clinical data call for robust and generalizable learning methods. To address these challenges\, there are several major directions I have explored: (i) (Robust) Machine Learning for Imperfect Medical Data: The development of machine learning models\, particularly in the context of label scarcity\, increasingly necessitates the collection of substantial annotated medical data. Moreover\, medical data often display a long-tailed class distribution\, which consequently results in notable imbalance issues. To this end\, there are several growing interests in training machine learning models jointly across imbalanced class distributions and limited annotations. I have developed novel\, efficient\, statistically consistent algorithms to improve empirical performance for biomedical image analysis. (ii) Learning with Theoretical Guarantees: As machine learning methods have become ubiquitous in clinical decision-making\, their reliability and interpretability have become important. This is particularly crucial in the field of biomedical image analysis\, where decision outcomes can have profound implications. I have developed novel machine learning algorithms that enable provably accurate anatomical modeling with theoretical guarantees. (iii) Generalize across Diverse Biomedical Data: The development of medical foundation models often requires massive and diverse biomedical data. To this end\, I have developed various foundation models for biomedical imaging data and explored novel applications of these models. I have also developed novel medical AI Agents that lead to the scalable and accurate predictive modeling\, particularly for distribution shift problems. \nSpeaker Bio\nChenyu You is an Assistant Professor in the Department of Applied Mathematics & Statistics and Department of Computer Science at Stony Brook University. He is also the core faculty member of the CVLab\, AI institute\, and affiliated with the Institute for Advanced Computational Science. His research focuses on both fundamental and applied problems in computer vision and machine learning\, often with a focus on generalization\, and making machine learning more reliable. Our applied research includes applications to healthcare\, biomedical imaging\, and cognitive neuroscience. He received his Ph.D. in 2024 from Yale University under the advisement of James S. Duncan\, his M.S. in 2019 from Stanford University under the advisement of Daniel Rubin\, and his B.S. in 2017 from Rensselaer Polytechnic Institute under the advisement of Ge Wang\, all in electrical engineering. He has also spent wonderful time at Facebook AI Research (FAIR)\, as well as Google Research. He serves on the Medical Image Computing and Computer-Assisted Intervention Society (MICCAI)\, and the SUNY AI Symposium Planning Committee\, and as associate editors for IEEE Transactions on Medical Imaging\, Medical Image Analysis\, IEEE Transactions on Neural Networks and Learning Systems\, Pattern Recognition\, and Transactions on Machine Learning Research. He has received AAAI’26 New Faculty Highlights\, CPAL’26 Rising Stars Award\, Tinker Research Grant Award\, Lambda Research Grant Award\, ICML’25 Oral Presentation Award\, EMBC’25 Top Paper Award\, MICCAI’25 NIH Registration Grant Award\, IEEE TMI’25 Distinguished Associate Editor Certificate of Excellence Award\, and Yale George P. O’Leary Graduate Fellowship\, and has been ranked as the World’s Top 2% most-cited scientists by Stanford University since 2024\, is a member of the Sigma Xi scientific research society\, and received the Excellence in Teaching Award for Spring and Fall 2025. For more information\, please check his website: https://chenyuyou.me/. \nHosted by: Professor Yuyin Zhou \nLocation: Engineering 2\, Room E2-180 (Refreshments such as fruit\, pastries\, coffee\, and tea will be provided.) \nZoom Option: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3
URL:https://events.ucsc.edu/event/cse-colloquium-robust-machine-learning-for-biomedical-data-efficiency-reliability-and-generalizability/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations,Seminars
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/03/BElogoWHITE.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260421T100000
DTEND;TZID=America/Los_Angeles:20260421T113000
DTSTAMP:20260428T120807
CREATED:20260401T234645Z
LAST-MODIFIED:20260401T234645Z
UID:10011845-1776765600-1776771000@events.ucsc.edu
SUMMARY:BE Climate & Cookies Student Pop-Up!
DESCRIPTION:Come get excited about Baskin Engineering Climate Week at our student pop-up! 🌎 \nClimate Week is a chance to explore how Baskin Engineering is addressing climate challenges through innovative research\, teaching\, and hands-on projects. \nDiscover the events happening throughout the week and find ways to get involved! \nSwing by for FREE BE swag\, coffee\, cookies\, Climate Week stickers\, and more—first come\, first served! \nWhere: BE Courtyard\nWhen: Tuesday\, April 21\, 10:00-11:30 a.m. \nWe hope to see you there!
URL:https://events.ucsc.edu/event/be-climate-week-pop-up-2026/
LOCATION:Jack Baskin Engineering\, Baskin Engineering 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Social Gathering
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/04/BE-climate-week-pop-up.png
GEO:37.000369;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Jack Baskin Engineering Baskin Engineering 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Baskin Engineering 1156 High Street:geo:-122.0632371,37.000369
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260415T173000
DTEND;TZID=America/Los_Angeles:20260415T203000
DTSTAMP:20260428T120807
CREATED:20260325T220453Z
LAST-MODIFIED:20260402T171331Z
UID:10011772-1776274200-1776285000@events.ucsc.edu
SUMMARY:Kraw Lecture: At the Forefront of AI: Innovation and Discovery
DESCRIPTION:Artificial intelligence is transforming how we understand and solve the world’s most complex challenges—while at the same time causing new challenges and concerns. We invite you to join us for a special UC Santa Cruz Kraw Lecture showcasing the faculty whose groundbreaking research in artificial intelligence is transforming science\, technology\, and society. From advances in autonomous systems and natural language processing to the development of sustainable and responsible AI\, this conversation will highlight the innovative work taking place across disciplines and the real-world impact it is poised to have. \nModerated by special guest Ahmad Thomas\, CEO of the Silicon Valley Leadership Group (SVLG)\, this dynamic discussion will bring together leading researchers to explore how these technologies are shaping the future—accelerating discovery\, addressing complex global challenges\, and opening new frontiers for collaboration. Gain insight into the ideas\, discoveries\, and collaborations shaping the next generation of artificial intelligence research and hear from the leaders advancing this work.\n \n\n\nIn-Person Reception: 5:30 p.m.\nLecture: 6:15 p.m.\n\nRegister Now
URL:https://events.ucsc.edu/event/kraw-lecture-at-the-forefront-of-ai-innovation-and-discovery/
LOCATION:The Quad Conference Center\, 2400 Sand Hill Rd\, Menlo Park\, CA\, 94025\, United States
CATEGORIES:Lectures & Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/03/2526-014E_Kraw_Lecture_banner-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260415T110000
DTEND;TZID=America/Los_Angeles:20260415T121500
DTSTAMP:20260428T120807
CREATED:20260407T155318Z
LAST-MODIFIED:20260407T155318Z
UID:10012050-1776250800-1776255300@events.ucsc.edu
SUMMARY:CSE Colloquium - Algorithmic Problems in Discrete Choice by Ravi Kumar
DESCRIPTION:Presenter: Ravi Kumar\, Google \nAbstract:\nIn discrete choice\, a user selects one option from a finite set of available alternatives\, a process that is crucial for recommendation systems applications in e-commerce\, social media\, search engines\, etc.  A popular way to model discrete choice is through Random Utility Models (RUMs).  RUMs assume that users assign values to options and choose the one with the highest value from among the available alternatives.  RUMs have become increasingly important in the Web era; they offer an elegant mathematical framework for researchers to model user choices and predict user behavior based on (possibly limited)  observations.   While RUMs have been extensively studied in behavioral economics and social sciences\, many basic algorithmic tasks remain poorly understood.  In this talk\, we will discuss various algorithmic and learning questions concerning RUMs. \nBio: \nRavi Kumar has been a research scientist at Google since 2012. Prior to this\, he was at the IBM Almaden Research Center and at Yahoo! Research. His interests include algorithms for massive data\, ML/privacy\, and the theory of computation. He maintains an extensive publication record that includes Test-of-Time Awards from STOC and WWW\, as well as Best Paper Awards from KDD and WWW\, to mention a few. He is an ACM fellow.\n\nHosted by: Professor Sungjin Im \n\nDate and Time: Wednesday\, April 15\, 2026 from 11:00 am – 12: 15 pm \nLocation: Engineering 2\, Room E2-180 (Refreshments such as fruit\, pastries\, coffee\, and tea will be provided.) \n\nZoom Option: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3 \n\n 
URL:https://events.ucsc.edu/event/cse-colloquium-algorithmic-problems-in-discrete-choice-by-ravi-kumar/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations,Seminars
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/03/BElogoWHITE.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260414T140000
DTEND;TZID=America/Los_Angeles:20260414T160000
DTSTAMP:20260428T120807
CREATED:20260326T162922Z
LAST-MODIFIED:20260326T162922Z
UID:10011787-1776175200-1776182400@events.ucsc.edu
SUMMARY:Castello\, J. (CSE) - Space Mission Simulation From the Outside In
DESCRIPTION:Robotic space missions often use discrete event simulation to reduce risk in operation. A simulation applies a set of planned activities to a model of mission resources\, and the model’s observed behavior is used to predict real-world outcomes. However\, logically concurrent activities are typically simulated under one possible linearization of their events – an order that may not reflect the eventual reality. Some simulation systems provide mechanisms for controlling the order of events; but this is a solution to a self-imposed problem. We instead question the assumption causing this problem: that events are totally ordered to begin with. \nWe study Merlin\, an open-source simulator developed at Caltech’s Jet Propulsion Laboratory and designed by the author\, that exchanges the traditional total order of events for a *partial* order. Under this approach\, a resumed activity can only observe events that causally precede its resumption\, and concurrent events are reconciled under custom policies. However\, the resulting design is more complex (and less understood) than that of linearizing simulators\, obscuring its key insights. As such\, we have developed Eidolon: a compact core calculus for Merlin-style simulation whose operational semantics follows an “outside-in”\, substitution-based execution model. Although Eidolon is derived from the concepts present in Merlin\, we intend it to be a vehicle for exploring non-linearizing simulation in general. \nFirst\, we propose making Eidolon *incremental*: a change to the set of planned activities should not incur a full resimulation from scratch except in the worse case\, instead reusing any cached computations that are not sensitive to the change. Since mission planning is a highly iterative process involving many simulations and subsequent tweaks to the plan\, incremental resimulation may allow plans to be finalized in less time\, or allow higher-quality plans to be obtained in the same amount of time. The substitution-oriented approach of Eidolon is what makes this feasible\, since individual computations align cleanly with subtree boundaries. \nSecond\, in the spirit of Reynolds’ defunctionalization and Danvy’s rational reconstruction\, we propose developing a denotational semantics for Eidolon and demonstrating its mechanical conversion into an abstract machine. As a mathematical artifact\, Eidolon is designed for reasoning and legibility rather than efficiency; nonetheless\, defunctionalization allows us to *refactor* our semantics into something that stands a chance of being practical. In particular\, defunctionalization reifies the recursive structure of a denotational semantics into an explicit data structure. As a result\, the defunctionalized form of Eidolon will recover an explicit priority queue like that of traditional linearizing simulators\, but without their assumption of total ordering. \nEvent Host: Jonathan Castello\, Ph.D. Student\, Computer Science and Engineering \nAdvisor: Lindsey Kuper  \nZoom- https://ucsc.zoom.us/j/98171466380?pwd=L2rkpr8tEt0MZamYbxxPTfvhAd4gl6.1 \nPasscode- 990848 \n 
URL:https://events.ucsc.edu/event/castello-j-cse-space-mission-simulation-from-the-outside-in/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/03/ph.d.-presentation-graphic-option-1.jpg
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260401T110000
DTEND;TZID=America/Los_Angeles:20260401T121500
DTSTAMP:20260428T120807
CREATED:20260325T164503Z
LAST-MODIFIED:20260330T203519Z
UID:10011765-1775041200-1775045700@events.ucsc.edu
SUMMARY:CSE Colloquium - Messages from across the event horizon:  AI Agentic Design for Computer Architecture (and more generalizable learnings)
DESCRIPTION:Presenter: Christopher Fletcher\, UC Berkeley \nAbstract: \nIt is difficult to escape the hype of agentic coding.  Is the hype real?  Are we still living in ~Summer 2025 — when AI coding would accomplish little more than upset its human supervisor?  Or has a level shift in technology finally arrived? \nIn this talk I will argue the latter.  I will describe a self-imposed experiment to discover modern AI coding tools’ capabilities (starting mid February 2026).  I will try (my best) to communicate my utter and sheer surprise at where the state of the art actually is.  Then I will do a deep dive and try to relay everything I have learned about this new engineering discipline—based on my attempts to push the technology as hard as I can for the past 1.5 months.  I will conclude by pontificating about the future of computer architecture and academic research more generally. \nBio: \nChristopher Fletcher is an Associate Professor of EECS at UC Berkeley. He is a computer architect whose research spans architecture\, security\, and domain-specific acceleration\, especially at their intersections from cryptography and hardware attacks to algorithm-to-hardware co-design. His work has received 31 paper recognitions and several other honors\, including the NSF CAREER Award\, Intel and Google faculty awards\, UIUC research and promotion awards\, election to DARPA ISAT\, and MIT’s George M. Sprowls Award\, with related work also recognized by Scientific American as one of ten “World Changing Ideas.” \nHosted by: Professor Alvaro Cardenas \nLocation: Engineering 2\, Room E2-180 (Refreshments such as fruit\, pastries\, coffee\, and tea will be provided.) \nZoom Option: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3 \n 
URL:https://events.ucsc.edu/event/cse-colloquium-messages-from-across-the-event-horizon-ai-agentic-design-for-computer-architecture-and-more-generalizable-learnings/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations,Seminars
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/03/BElogoWHITE.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260313T140000
DTEND;TZID=America/Los_Angeles:20260313T150000
DTSTAMP:20260428T120807
CREATED:20260219T170502Z
LAST-MODIFIED:20260219T170502Z
UID:10009254-1773410400-1773414000@events.ucsc.edu
SUMMARY:Wang\, H. (CSE) - Accelerating RTL Simulation with Specialized Graph Partitioners
DESCRIPTION:Register transfer level (RTL) simulation is an invaluable tool for developing\, debugging\, verifying\, and validating hardware designs. However\, the performance of RTL simulation has long been a limiting factor in industry. Despite the inherent parallelism of hardware\, current RTL simulators have not achieved practical performance gains due to fundamental challenges in communication\, synchronization\, memory bandwidth\, and architectural mapping. \nThis dissertation addresses the RTL simulation performance problem from three complementary perspectives: optimizing simulation latency through parallelism\, improving aggregate throughput via deduplication\, and enabling efficient GPU acceleration with RTL-native semantics. \nFirst\, we present RepCut\, a parallel RTL simulation methodology that uses replication-aided partitioning to cut circuits into balanced partitions with minimal overlaps. By replicating the overlaps\, RepCut eliminates problematic data dependences between partitions and significantly reduces synchronization overhead. RepCut achieves superlinear speedups of up to 27.10x using 24 threads with only a 3.81% replication cost. \nSecond\, we introduce Simulation Deduplication\, a technique that exploits the extensive reuse of building blocks in modern hardware designs. By generating shared code for duplicated instances and carefully co-scheduling their execution\, we reduce the instruction cache footprint and memory bandwidth pressure. This approach achieves up to 1.95x speedup for single simulations and 2.09x improvement in overall batch simulation throughput. \nThird\, we present Toucan\, a GPU-accelerated RTL simulation framework that preserves RTL semantics rather than flattening designs to gate-level netlists. By leveraging native GPU arithmetic operations and introducing warp-level micro-partitioning with shuffle-based communication\, Toucan achieves efficient mapping of irregular circuit topologies to GPU SIMT architectures while maintaining fast compilation times. Toucan achieves up to 4.73x speedup over the state-of-the-art GPU RTL simulator on large multi-core designs. \nTogether\, these three approaches provide a comprehensive solution to RTL simulation performance optimization\, demonstrating significant improvements over state-of-the-art commercial and open-source simulators across multiple hardware platforms and design scales. \nEvent Host: Haoyuan Wang\, Ph.D. Candidate\, Computer Science and Engineering \nAdvisor: Jose Renau \nZoom- https://ucsc.zoom.us/j/94044618343?pwd=xZkK8GmD28P2Vf8pbyl6aoOaNxxhya.1 \nPasscode- 574772
URL:https://events.ucsc.edu/event/wang-h-cse-accelerating-rtl-simulation-with-specialized-graph-partitioners/
LOCATION:Jack Baskin Engineering\, Baskin Engineering 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/02/ph.d.-presentation-graphic-option2.jpg
GEO:37.000369;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Jack Baskin Engineering Baskin Engineering 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Baskin Engineering 1156 High Street:geo:-122.0632371,37.000369
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260313T100000
DTEND;TZID=America/Los_Angeles:20260313T120000
DTSTAMP:20260428T120807
CREATED:20260304T172425Z
LAST-MODIFIED:20260304T172425Z
UID:10009393-1773396000-1773403200@events.ucsc.edu
SUMMARY:Moghadam\, M. (CE) - Constraint-Aware Scene Understanding and Trajectory Generation Using Deep Reinforcement Learning for Autonomous Vehicles
DESCRIPTION:Advanced driver-assistance systems (ADAS) are commonly organized as modular pipelines that transform raw sensor measurements into low-level actuation commands through perception\, planning\, and control. While learning-based methods have achieved state-of-the-art performance in perception and environment modeling\, the planning layer remains a key bottleneck for reliable autonomy. Highway driving in particular requires long-horizon reasoning and socially aware interaction with multiple actors\, while also producing smooth and dynamically feasible motion that can be tracked by classical controllers. \nThis thesis focuses on scene understanding and planning for highway driving. We study the problem through two complementary simulation environments: the high-fidelity CARLA simulator for motion planning and continuous trajectory generation under realistic vehicle dynamics and road geometry\, and the lightweight HighwayEnv simulator for interaction-rich behavior planning at high episode throughput. \nWe present three planning contributions that increase autonomy. First\, we introduce a modular hierarchical planning framework in Frenet space that combines long-term decision-making with short-term trajectory optimization. The approach includes a corridor-based dynamic obstacle avoidance strategy that generates spatiotemporal polynomial trajectories and supports diverse driving styles through interpretable parameter tuning. Second\, we propose an end-to-end continuous deep reinforcement learning approach that unifies decision-making and motion planning into a single policy that outputs continuous polynomial trajectories in the Frenet frame. A spatiotemporal observation tensor and a temporal convolutional backbone enable the learned planner to exploit interaction history and outperform optimization-based and discrete RL baselines in CARLA. Third\, we develop an interaction-aware behavior planning neural network architecture that couples trajectory prediction with high-level decision-making via a social pooling scene encoder built on actor histories and an ego-centered BEV representation. This unified design improves RL social awareness\, safety\, and overall driving performance in multi-agent highway scenarios in HighwayEnv. \nAcross extensive simulation studies\, the results show that constraint-aware representations and learning-based policies can improve planning quality beyond hand-crafted objectives\, especially when the policy is equipped with spatiotemporal social context while retaining classical feedback control for stable trajectory tracking. Finally\, we provide supporting simulation and evaluation infrastructure\, including observation tensor and neural network designs\, BEV utilities\, and scalable training and testing pipelines\, to enable reproducible research on learning-based planning in interactive traffic. \nEvent Host: Majid Moghadam\, Ph.D. Candidate\, Computer Engineering  \nAdvisor: Gabriel Elkaim \nZoom- https://ucsc.zoom.us/j/95848602314?pwd=2jlktZ6BChlXcyqT3anX4ZuKrYV4wE.1 \nPasscode- 325939
URL:https://events.ucsc.edu/event/moghadam-m-ce-constraint-aware-scene-understanding-and-trajectory-generation-using-deep-reinforcement-learning-for-autonomous-vehicles/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/02/ph.d.-presentation-graphic-option-1.jpg
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260313T093000
DTEND;TZID=America/Los_Angeles:20260313T110000
DTSTAMP:20260428T120807
CREATED:20260217T203948Z
LAST-MODIFIED:20260217T203948Z
UID:10009241-1773394200-1773399600@events.ucsc.edu
SUMMARY:Fan\, Y. (CSE) - Building Human-Centered Multimodal AI Agents
DESCRIPTION:As multimodal artificial intelligence systems become increasingly embedded in everyday technology\, there is a growing need to design human-centered AI agents that support and amplify human capabilities rather than replace them. This dissertation investigates how to build human-centered multimodal AI agents\, framing human-centeredness as an agent-level objective that requires both accessible\, assistive interaction and reliable\, trustworthy behavior across physical and digital environments. This dissertation explores two complementary dimensions of human-centered agent design. The first focuses on enhancing accessibility through conversational and interactive agents that assist users in everyday tasks. We study both embodied and digital settings in which agents reduce physical and cognitive burdens via natural language interaction\, including hands-free drone control\, navigation assistance in unfamiliar environments\, and interactive access to complex graphical user interfaces. The second dimension focuses on strengthening agent capability to improve reliability and trust. We investigate how agents can acquire environment-specific knowledge through autonomous exploration and how they can reason about visual information in a grounded and transparent manner\, drawing inspiration from human learning and reasoning behaviors. \nEvent Host: Yue Fan\, Ph.D. Candidate\, Computer Science and Engineering \nAdvisor: Xin Eric Wang \nZoom- https://ucsc.zoom.us/j/99619642071?pwd=dwWOlkJxjbamgpB4IbRxYDXbngqXOE.1 \nPasscode- 467959
URL:https://events.ucsc.edu/event/fan-y-cse-building-human-centered-multimodal-ai-agents/
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/02/ph.d.-presentation-graphic-option2.jpg
LOCATION:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260311T110000
DTEND;TZID=America/Los_Angeles:20260311T130000
DTSTAMP:20260428T120807
CREATED:20260311T160620Z
LAST-MODIFIED:20260311T160620Z
UID:10011304-1773226800-1773234000@events.ucsc.edu
SUMMARY:Yang\, S. (CSE) - Beyond Image Editing: Building Generalized Image Customization Systems
DESCRIPTION:Current generative vision models struggle with image customization that requires multi-step reasoning or real-world knowledge. This proposal introduces generalized image customization\, enabling systems to execute complex\, inferential modifications rather than just simple edits. The research focuses on the foundational framework required for this generalization\, specifically high-quality training data\, scalable evaluation benchmarks\, self-improving training paradigms that reduce reliance on paired annotations\, and unified multi-modal architectures. Building on two completed studies in data quality and evaluation\, this proposal outlines two future research directions to develop capable\, annotation-efficient\, and reasoning-native image customization systems. \nEvent Host: Siwei Yang\, Ph.D. Student\, Computer Science and Engineering \nAdvisor: Cihang Xie \nZoom- https://ucsc.zoom.us/j/3852138080?pwd=Z0MyTVM2WjdCbEM4OXVxWUhhei84dz09 \n 
URL:https://events.ucsc.edu/event/yang-s-cse-beyond-image-editing-building-generalized-image-customization-systems/
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/02/ph.d.-presentation-graphic-option-1.jpg
LOCATION:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260311T110000
DTEND;TZID=America/Los_Angeles:20260311T121500
DTSTAMP:20260428T120807
CREATED:20260303T181914Z
LAST-MODIFIED:20260303T181914Z
UID:10009389-1773226800-1773231300@events.ucsc.edu
SUMMARY:CSE Colloquium: Co-Active AI-Assisted Programming
DESCRIPTION:Presenter: Nadia Polikarpova\, UCSD \nAbstract: \nAI-assisted programming has rapidly moved from novelty to default. Today\, most developers use AI coding tools\, and increasingly rely on agentic systems capable of making multi-step design and implementation decisions with minimal human guidance. While these systems boost productivity\, they also introduce new risks: developers may disengage from the reasoning behind generated code\, leading to shallow understanding\, loss of ownership\, and what is increasingly described as cognitive debt. \nIn this talk\, I argue that AI-driven software development must be co-active: humans and AI should remain continuously engaged in a shared process of understanding and decision-making. I will present two complementary research directions toward this goal. The first focuses on observability—helping developers understand\, validate\, and trace the behavior of AI-generated code. The second focuses on controllability—making AI decisions explicit\, persistent\, and steerable. Together\, these ideas restore programmer agency while maintaining the productivity gains of AI-assisted development. \nBio: \nNadia Polikarpova is an associate professor at UC San Diego\, and a member of the Programming Systems group. She received her Ph.D. in Computer Science from ETH Zurich in 2014\, and then spent some time as a postdoctoral researcher at MIT. Nadia’s research interests are at the intersection of programming languages\, AI\, human-computer interaction\, and social computing. \nHosted by: Professor Nikos Tziavelis \nLocation: Engineering 2\, Room E2-180 (*Refreshments such as coffee\, tea\, pastries\, and fresh fruit will be available.) \nZoom: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3
URL:https://events.ucsc.edu/event/cse-colloquium-co-active-ai-assisted-programming/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations,Seminars
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/02/BElogoWHITE.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260310T163000
DTEND;TZID=America/Los_Angeles:20260310T173000
DTSTAMP:20260428T120807
CREATED:20260217T184921Z
LAST-MODIFIED:20260217T184921Z
UID:10009240-1773160200-1773163800@events.ucsc.edu
SUMMARY:Mashhadi\, N. (CSE) - Compositional\, Clinically Conditioned\, and Confound-Aware Deep Learning for Alzheimer’s Disease Neuroimaging
DESCRIPTION:Alzheimer’s disease (AD) is a progressive neurodegenerative disorder and a leading cause of dementia. Neuroimaging and clinical biomarkers can reveal early disease changes\, but building reliable machine learning models is difficult because data come from different scanners and sites\, some modalities are missing\, labeled cohorts are limited\, and factors such as age and scanner/site effects can bias results. \nThis dissertation develops machine learning methods for robust\, interpretable\, and controllable analysis of AD-related neuroimaging data. First\, I introduce a modular\, graph-based framework for multimodal AD detection that treats datasets and models as nodes and directed edges that can be combined to build more complex predictors. Second\, I propose a clinically conditioned 3D VAE-GAN to synthesize brain MRI\, enhanced with diffusion-driven sampling in clinical feature space to improve realism and control\, supporting data augmentation. Third\, I present a disentangled 3D masked autoencoder (MAE) that learns separated representations for age\, pathology\, and scanner effects\, making it possible to isolate and adjust age\, pathology\, or scanner effects\, while remaining reliable across sites. \nTogether\, these contributions advance practical methods for modular prediction\, controllable image generation\, and confound-aware representation learning in neuroimaging\, with an emphasis on generalization and interpretability for clinically relevant applications. \nEvent Host: Najmeh Mashhadi\, Ph.D. Candidate\, Computer Science and Engineering \nAdvisor: Razvan Marinescu \nZoom- https://ucsc.zoom.us/j/98195204428?pwd=nyfvbmd9t81Xj5Z3yPPVtu4R58CXHq.1 \nPasscode- 688069
URL:https://events.ucsc.edu/event/mashhadi-n-cse-compositional-clinically-conditioned-and-confound-aware-deep-learning-for-alzheimers-disease-neuroimaging/
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/02/ph.d.-presentation-graphic-option2.jpg
LOCATION:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260309T140000
DTEND;TZID=America/Los_Angeles:20260309T160000
DTSTAMP:20260428T120807
CREATED:20260303T174216Z
LAST-MODIFIED:20260303T174216Z
UID:10009384-1773064800-1773072000@events.ucsc.edu
SUMMARY:Harrison\, D. (CS) - Multi-Level Control in Neural Dialogue Generation: Style\, Semantics\, and Selection through Over-Generation and Ranking
DESCRIPTION:End-to-end neural generation models have largely displaced the modular architectures that once gave dialogue system designers explicit control over what is said and how it is said. While these models produce fluent text\, they collapse content planning\, sentence planning\, and surface realization into a single undifferentiated decoding step\, sacrificing the controllable structure that earlier systems provided. This dissertation investigates how that structure can be recovered through the over-generate-and-rank (OGR) paradigm: generating multiple candidate outputs and selecting among them using learned or prompt-based ranking functions that jointly optimize semantic fidelity\, stylistic appropriateness\, and conversational coherence. We instantiate OGR at three levels of natural language generation for dialogue: utterance-level stylistic control\, cross-domain semantic evaluation\, and dialogue-level response selection. \nFirst\, we show that explicit conditioning mechanisms\, specifically decoder-level side constraints for personality variation and discourse contrast\, re-introduce stylistic control into neural sequence-to-sequence models without compromising semantic accuracy. Second\, we demonstrate that prompt-based learning with structured linguistic profiles achieves near-perfect personality accuracy and effectively zero slot error rate when combined with ranking\, establishing that LLM prompting with explicit pragmatic specifications can match or exceed fine-tuning for personality-conditioned generation. Third\, we develop a cross-domain semantic error rate evaluation framework that frames slot error computation as an extraction task\, using a LoRA-adapted language model to extract meaning representations from generated text and a trained ranker to select among candidate extractions\, achieving reliable evaluation across 23 topic domains without domain-specific rules. Fourth\, we build and evaluate a speaker-aware transformer response ranker for Athena\, our Alexa Prize socialbot\, demonstrating that learned ranking over heterogeneous generator pools produces significantly longer conversations and higher user ratings than heuristic rule-based selection in a live A/B study with over 6\,000 conversations. \nA unifying finding emerges across all four contributions: the pragmatic features that control personality style in generation—acknowledgements\, engagement questions\, hedges\, exclamations—are the same features that distinguish high-quality from mediocre responses in open-domain dialogue. This parallel reveals that stylistic control and response ranking are complementary mechanisms for achieving the same goal: making dialogue systems sound more natural and engaging. Together\, these results support the dissertation’s central hypothesis that over-generate-and-rank provides a general\, extensible mechanism for controllable neural language generation\, restoring explicit decision points where competing communicative objectives can be weighed. The ranking function serves a role analogous to the sentence planner in classical NLG architectures\, but operates on the outputs of modern neural and LLM-based generators. \n  \nEvent Host: Davan Harrison\, Ph.D. Candidate\, Computer Science \nAdvisor: Marilyn Walker \n 
URL:https://events.ucsc.edu/event/harrison-d-cs-multi-level-control-in-neural-dialogue-generation-style-semantics-and-selection-through-over-generation-and-ranking/
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/02/ph.d.-presentation-graphic-option2.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260305T130000
DTEND;TZID=America/Los_Angeles:20260305T150000
DTSTAMP:20260428T120807
CREATED:20260217T182432Z
LAST-MODIFIED:20260217T182432Z
UID:10009238-1772715600-1772722800@events.ucsc.edu
SUMMARY:Xu\, Y. (CSE) - Right Place\, Right Time: Accelerating Edge Computation on Modern Heterogeneous SoCs
DESCRIPTION:Modern edge computing increasingly relies on heterogeneous System-on-Chip (SoC) architectures. These chips tightly integrate general-purpose CPUs with various specialized accelerators\, including GPUs\, FPGAs\, and AI accelerators\, all under a shared memory architecture. Although these shared-memory SoCs enable more efficient communication and data sharing between different processing units\, they are notoriously difficult to program and tune due to architectural diversity across vendors and asymmetric compute capabilities within each SoC. \nThis dissertation introduces Redwood and BetterTogether\, two frameworks that rethink CPU-accelerator collaboration on heterogeneous SoCs. Redwood targets a class of algorithms termed traverse–compute\, that combine irregular tree traversals with dense leaf-level computation\, e.g.\, Nearest-Neighbor Search and Barnes–Hut algorithm. \nIt addresses the efficient mapping of these algorithms onto heterogeneous systems by exploiting the architectural strengths of CPUs\, GPUs\, and FPGAs. BetterTogether extends this methodology to a different class of edge workloads\, specifically multi-stage pipelines and neural networks commonly used in computer vision tasks. Furthermore\, it introduces interference-aware analysis and scheduling techniques tailored for mobile SoCs. Finally\, to broaden the scope of heterogeneous acceleration\, we evaluated emerging domain-specific accelerators. We provide a preliminary analysis of Tensor Processing Units and Tensor Cores within the context of modern programming abstractions. \nEvent Host: Yanwen Xu\, Ph.D. Candidate\, Computer Science and Engineering \nAdvisor: Tyler Sorensen \nZoom- https://ucsc.zoom.us/j/5354629158?pwd=0CVhbwLuXDMX5fAGZd63tcfNqDWp0t.1 \nPasscode- 114514
URL:https://events.ucsc.edu/event/xu-y-cse-right-place-right-time-accelerating-edge-computation-on-modern-heterogeneous-socs/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/02/ph.d.-presentation-graphic-option2.jpg
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260304T110000
DTEND;TZID=America/Los_Angeles:20260304T121500
DTSTAMP:20260428T120807
CREATED:20260217T182353Z
LAST-MODIFIED:20260217T182353Z
UID:10009237-1772622000-1772626500@events.ucsc.edu
SUMMARY:CSE Colloquium - Improving Efficiency and Reliability of Foundation Models in Clinical AI
DESCRIPTION:Presenter: Vasiliki “Vicky” Bikia\, PhD\, Stanford Department of Biomedical Data Science and Institute for Human-Centered AI (HAI) \nAbstract: \nDeploying foundation models in health requires both computational efficiency and reliable generation. In this talk\, I present two studies that address these dimensions separately but with a shared goal of real-world clinical deployment. The first study focuses on reduced-resolution distillation for multimodal clinical data\, particularly medical imaging. As model and input sizes increase\, inference cost and memory constraints become major barriers to deployment. We investigate how high-capacity teacher models can transfer structured knowledge to compact student models trained on downsampled images\, using embedding-space supervision to preserve clinically meaningful representations while reducing computational footprint. The second study examines the reliability of AI-generated clinical text. Foundation models are increasingly used to produce discharge summaries and patient-facing explanations\, yet fluency does not guarantee safety. We develop a structured evaluation framework grounded in clinical error taxonomies and clinician-calibrated metrics to quantify hallucinations\, omissions\, and semantic misalignment. Together\, these studies emphasize that scalable clinical AI requires not only smaller and faster models\, but also rigorous evaluation of generative reliability before deployment. \nBio: \nVasiliki Bikia is a Postdoctoral Researcher at Stanford University\, affiliated with the Department of Biomedical Data Science and the Stanford Institute for Human-Centered Artificial Intelligence (HAI). She received an Advanced Diploma in Electrical and Computer Engineering from the Aristotle University of Thessaloniki\, and a Ph.D. in Bioengineering from the Swiss Federal Institute of Technology in Lausanne (EPFL). Her research focuses on medical foundation models\, structured representations of health data\, and the evaluation of generative systems in clinical settings. Previously\, she was a Machine Learning Scientist at the Mussallem Center for Biodesign at Stanford University\, where she developed software pipelines to improve data accessibility and interoperability in digital health applications. Vasiliki was selected as an MIT Rising Star in EECS (2025) and as an Emerson Consequential Scholar (2025)\, and is actively engaged with the Silicon Valley entrepreneurial ecosystem through collaborations at the intersection of research\, industry\, and healthcare. She is an organizing member of the Conference on Health\, Inference\, and Learning (CHIL) and serves as Unconference Chair for the 2025 and 2026 editions\, where she leads the design and execution of the entrepreneurship-focused track bridging academic research and real-world deployment. Her work has appeared in venues including IEEE journals\, npj Digital Medicine\, Nature Communications\, and leading AI conferences\, and she has contributed to multiple funded research proposals and clinical studies at the intersection of AI\, medicine\, and translational impact. \nHosted by: Professor Nikos Tziavelis \nLocation: Engineering 2\, E2-180 (*Refreshments such as coffee\, tea\, fresh fruit\, and pastries will be provided) \nZoom: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3
URL:https://events.ucsc.edu/event/cse-colloquium-improving-efficiency-and-reliability-of-foundation-models-in-clinical-ai/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations,Seminars
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/02/Screenshot-2026-02-17-at-10.19.22-AM.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260227T094500
DTEND;TZID=America/Los_Angeles:20260227T160000
DTSTAMP:20260428T120807
CREATED:20260126T234626Z
LAST-MODIFIED:20260225T003412Z
UID:10009117-1772185500-1772208000@events.ucsc.edu
SUMMARY:Semiconductor Career Summit - From Campus to Silicon Valley
DESCRIPTION:A SEMI Professional Development Seminar organized by the SEMI Silicon Valley Chapter – Connecting College Students to the Semiconductor Industry. Learn about career opportunities in high tech and acquire valuable\, practical information that will help you choose career directions and plan for your success. \nCome learn about careers in the semiconductor industry at the SEMI Professional Development Seminar hosted by UC Santa Cruz. \n\nListen to professionals in the industry talk about their roles and find out how to prepare for jobs in the Semiconductors Industry.\nDiscover semiconductor job opportunities you didn’t know existed (internship and entry-level) and how you can prepare for them through job searches\, interviews\, resumes\, and more.\nMeet with professionals and executives during our speed mentoring\, mock interview\, and networking sessions.\n\nAll majors are welcome! Students with a background in Engineering\, Computer Science\, Chemistry\, Physics\, Math\, Data Science\, and Business are strongly encouraged to attend. \n\nEnjoy free food\, free swag\, and giveaways.\nStudents can come and go.\n\nEVENT is FREE but registration is required. Register by Feb 20th to secure a lunch.  \nEvent is organized by SEMI in collaboration with Career Success\, Baskin Engineering and the Innovation & Business Engagement Hub. \nYou Belong Here: The programs and services described here are open to all\, consistent with state and federal law\, as well as the University of California’s nondiscrimination policies. Every initiative—whether a student service\, faculty program\, or community event—is designed to be accessible\, inclusive\, and respectful of all identities. \nTo learn more\, please visit UC Nondiscrimination Statement or Nondiscrimination Policy for UC Publications. \nQuestions? Send to csuccess@ucsc.edu or visit Career Success at Hahn 125 East Entrance\nNeed accessibility support? Let us know at slugtalent@ucsc.edu at least two weeks prior to the event date.
URL:https://events.ucsc.edu/event/semiconductor-career-summit-from-campus-to-silicon-valley/
LOCATION:Stevenson Event Center\, Stevenson Service Road\, Santa Cruz\, CA\, 95064
CATEGORIES:Exhibits,Lectures & Presentations,Meetings & Conferences,Training
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/01/Screenshot-2026-02-11-at-12.47.54-PM.png
GEO:36.996897;-122.0512963
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Stevenson Event Center Stevenson Service Road Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Stevenson Service Road:geo:-122.0512963,36.996897
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260226T120000
DTEND;TZID=America/Los_Angeles:20260226T140000
DTSTAMP:20260428T120807
CREATED:20260129T143555Z
LAST-MODIFIED:20260223T200828Z
UID:10009134-1772107200-1772114400@events.ucsc.edu
SUMMARY:BE Club Bash - Engineers Week
DESCRIPTION:Discover innovation at the Baskin Engineering Club Bash\, an event celebrating National Engineers Week! \nMark your calendars for Thursday\, February 26\, 12–2 PM in the BE Courtyard! The BE Club Bash brings together student organizations across all engineering disciplines to showcase their projects\, demos\, and interactive activities. \nStop by to: \n\nExplore hands-on booths and demonstrations from student organizations\nLearn about engineering opportunities on campus and how to get involved\nChat with student leaders and hear about their experiences\nEnter our 3D printer raffle (must be present to win!)\nGrab snacks and BE swag while you explore\n\nThis is a great way to connect with the engineering community\, discover new ideas\, and have fun. We hope to see you there! RSVP here.
URL:https://events.ucsc.edu/event/be-club-bash-engineers-week/
LOCATION:Jack Baskin Engineering\, Baskin Engineering 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Social Gathering,Undergraduate
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/01/5-22-25-Slugworks-CL-049-3-scaled.jpg
GEO:37.000369;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Jack Baskin Engineering Baskin Engineering 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Baskin Engineering 1156 High Street:geo:-122.0632371,37.000369
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260225T173000
DTEND;TZID=America/Los_Angeles:20260225T190000
DTSTAMP:20260428T120807
CREATED:20260130T054047Z
LAST-MODIFIED:20260209T232119Z
UID:10009139-1772040600-1772046000@events.ucsc.edu
SUMMARY:Exploring Research Pathways at Baskin Engineering
DESCRIPTION:Curious how being part of a research lab can supercharge your experience as a Baskin Engineer?   \nJoin us for this informative event to learn about opportunities to solve open-ended problems\, build deeper technical skills\, and learn how to think like an engineer. \nWe’ll kick things off with a quick overview of the kinds of research opportunities available to undergrads and how to get started\, then you’ll hear directly from students who’ve worked in research labs as undergraduates. They’ll share what they actually did day-to-day\, the skills they built (technical and professional)\, and how research shaped their confidence\, career goals\, and next steps. We’ll then have pizza and networking to end the evening. \nWhether you’re aiming for industry\, graduate school\, or just want hands-on experience that goes beyond coursework\, this panel will help you understand how undergraduate research can set you apart—academically\, professionally\, and personally! \n\nRegister via Handshake. \nYOU BELONG HERE\nPrograms and services are open to all\, consistent with state and federal law\, as well as the University of California’s nondiscrimination policies. Every initiative—whether a student service\, faculty program\, or community event—is designed to be accessible\, inclusive\, and respectful of all identities. To learn more\, please visit UC Nondiscrimination Statement or Nondiscrimination Policy for UC Publications.
URL:https://events.ucsc.edu/event/exploring-research-pathways-at-baskin-engineering/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Seminars
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/01/BElogoWHITE.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260225T110000
DTEND;TZID=America/Los_Angeles:20260225T121500
DTSTAMP:20260428T120807
CREATED:20260224T172405Z
LAST-MODIFIED:20260224T172405Z
UID:10009274-1772017200-1772021700@events.ucsc.edu
SUMMARY:CSE Colloquium: Agile and evolvable software construction in the era of rapidly evolving hardware accelerator designs
DESCRIPTION:Presenter\n\nCharith Mendis\, Siebel School of Computing and Data Science\, University of Illinois at Urbana-Champaign\n\nAbstract\n\nModern AI workloads have become exceedingly abundant and important in the current computing landscape. As a result\, there have been numerous software and hardware innovations aimed at accelerating these workloads. However\, we observe a subtle disconnect between the software and hardware communities. Most software innovations target well-established hardware platforms such as CPUs (e.g.\, x86\, ARM) and GPUs (e.g.\, NVidia GPUs)\, while hardware innovations produce plenty of other tensor accelerator designs (e.g.\, Gemmini\, Feather\, Trainium) each year.\n\nWe asked the question\, why aren’t the software community using these accelerators or even evaluating on them? The simple yet undeniable reason is the lack of standardized software tooling compared to CPUs and GPUs. For an architecture to be used\, properly designed compiler backends\, correctness\, and performance testing tools should be abundant (e.g.\, CUDA ecosystem).\n\nIn this talk\, I will describe how we bridge this gap by automatically generating the necessary software tools for a large class of accelerators through the Accelerator Compiler Toolkit (ACT) ecosystem. Central to ACT is an ISA definition language\, TAIDL\, that for the first time standardizes the hardware-software interfaces for a large class of accelerators. Departing from the traditional approach of manually constructing test oracles\, performance models\, or retargetable compiler backends\, we instead introduce agile and evolvable methodologies to automatically generate such necessary tooling using both formal methods and machine learning techniques for any TAIDL-defined accelerator interface. I will show how such automation enables rapid software prototyping\, making rapidly evolving accelerator designs usable by the software community.\n\nBio\n\nCharith Mendis is an Assistant Professor in the Siebel School of Computing and Data Science at the University of Illinois at Urbana-Champaign. His broad research interests are at the intersection of compilers\, programming languages\, and machine learning. He received his Ph.D. and Master’s from the Massachusetts Institute of Technology and his B.Sc. from the University of Moratuwa. He is the recipient of the DARPA Young Faculty Award\, the NSF CAREER Award\, the Google ML and Systems Junior Faculty Award\, the Outstanding Advisor award at UIUC\, the William A. Martin Outstanding Master’s Thesis Award at MIT\, and the University Gold Medal for his B.Sc. He has won numerous paper awards\, including a Distinguished Paper Award at POPL\, a Best Student Paper Award at the IEEE BigData conference\, an honorable mention for the Best Artifact Award at SIGMOD\, a Best Paper Award at ML for Systems workshop at ISCA\, and an IEEE Top Picks Honorable Mention.\n\nHosted by: Professor Nikos Tziavelis\n\nLocation: Engineering 2\, E2-180 (Refreshments such as fruit\, pastries\, tea\, and coffee will be available for guests.)\n\nZoom: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3 
URL:https://events.ucsc.edu/event/cse-colloquium-agile-and-evolvable-software-construction-in-the-era-of-rapidly-evolving-hardware-accelerator-designs/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/02/BElogoWHITE.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260225T090000
DTEND;TZID=America/Los_Angeles:20260225T120000
DTSTAMP:20260428T120807
CREATED:20260210T221905Z
LAST-MODIFIED:20260210T221905Z
UID:10009196-1772010000-1772020800@events.ucsc.edu
SUMMARY:Liu\, C. (CSE) - Enabling LLM Unlearning at Inference Time by Decomposing Detection and Intervention
DESCRIPTION:Machine unlearning addresses the “right to be forgotten” under GDPR and enables privacy\, copyright\, and safety compliance in large language models. Training-based unlearning can remove targeted behavior on benchmarks\, but it scales poorly\, can degrade utility\, and can fail under adversarial prompting that recovers supposedly forgotten content. This prospectus proposes inference-time behavioral unlearning: rather than modifying weights to “erase” knowledge\, we detect when a query targets forgotten content and intervene in generation so the system behaves like a model never trained on that content. We formalize this approach as Detect-Intervene Decomposition and instantiate it with three complementary methods operating at the embedding\, token\, and reasoning levels under different access capabilities. Comprehensive experiments across entity unlearning\, hazardous knowledge removal\, and copyright protection demonstrate that our methods match or exceed training-based approaches while being orders of magnitude faster and preserving model utility. As LLMs increasingly operate as services with restricted weight access\, inference-time unlearning provides the only practical path for responsible AI deployment that respects privacy\, safety\, and legal requirements. \nEvent Host: Chris Liu\, Ph.D. Student\, Computer Science and Engineering \nAdvisor: Yang Liu \nZoom – https://ucsc.zoom.us/j/94799852992?pwd=EBFQe4U2lRNro1oJ8F36bgORhT2xSv.1 \nPasscode –  242384
URL:https://events.ucsc.edu/event/liu-c-cse-enabling-llm-unlearning-at-inference-time-by-decomposing-detection-and-intervention/
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/01/ph.d.-presentation-graphic-option-1-1.jpg
LOCATION:
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260224T170000
DTEND;TZID=America/Los_Angeles:20260224T181500
DTSTAMP:20260428T120807
CREATED:20260130T054112Z
LAST-MODIFIED:20260209T231917Z
UID:10009138-1771952400-1771956900@events.ucsc.edu
SUMMARY:AI and Security 101
DESCRIPTION:Join us for an informative conversation with Neta Haiby\, Head of Product | AI Security at Microsoft! \nArtificial Intelligence is transforming both cyber defense and cyber offense. It creates unique risks in how we build\, deploy\, and operate AI apps and Agents. This session examines how AI can be attacked or misused – through techniques such as jailbreaks\, intent breaking\, and supply-chain compromise and discusses practical defense strategies\, including guardrails\, access controls\, monitoring\, and evaluation. \nDesigned for students interested in cybersecurity and AI\, this session emphasizes a practical understanding of AI security. \nAttendees will also receive resources to help them further explore and get started in the field! \nDon’t miss this highly informative event! \nYOU BELONG HERE\nPrograms and services are open to all\, consistent with state and federal law\, as well as the University of California’s nondiscrimination policies. Every initiative—whether a student service\, faculty program\, or community event—is designed to be accessible\, inclusive\, and respectful of all identities. To learn more\, please visit UC Nondiscrimination Statement or Nondiscrimination Policy for UC Publications.
URL:https://events.ucsc.edu/event/ai-and-security-101/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Undergraduate,Workshop
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/01/Neta-Haiby.jpeg
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260224T103000
DTEND;TZID=America/Los_Angeles:20260224T113000
DTSTAMP:20260428T120807
CREATED:20260129T145348Z
LAST-MODIFIED:20260209T232106Z
UID:10009135-1771929000-1771932600@events.ucsc.edu
SUMMARY:Transform Your Future Pop-Up (Cookies Included!)
DESCRIPTION:Join Baskin Engineering to celebrate National Engineers Week with a sweet stop at the Transform Your Future Pop-Up (Cookies Included!) 🍪☕ \nThis year’s Engineers Week theme\, Transform Your Future\, is a powerful reminder that engineering doesn’t just shape our world—it shapes our opportunities\, our communities\, and the futures we can imagine for ourselves. \nSwing by the BE Courtyard to grab cookies\, coffee\, and BE swag (first come\, first served!) and take a moment to celebrate how you are transforming your future. \n📅 Date: Tuesday\, February 24⏰ Time: 10:30 a.m.📍 Location: BE Courtyard \nWe hope to see you there!
URL:https://events.ucsc.edu/event/transform-your-future-pop-up-cookies-included/
LOCATION:Jack Baskin Engineering\, Baskin Engineering 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Social Gathering,Undergraduate
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/01/ChristineLaPhotography-CA-UCSantaCruz-StudentLife-Day1-04092025-02192-1-scaled.jpg
GEO:37.000369;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Jack Baskin Engineering Baskin Engineering 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Baskin Engineering 1156 High Street:geo:-122.0632371,37.000369
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260220T140000
DTEND;TZID=America/Los_Angeles:20260220T160000
DTSTAMP:20260428T120807
CREATED:20260210T193542Z
LAST-MODIFIED:20260210T193542Z
UID:10009193-1771596000-1771603200@events.ucsc.edu
SUMMARY:Fredrickson\, K. (CSE) - Practical Anonymity with Formal Resistance to Traffic Analysis
DESCRIPTION:Anonymous communication systems hide who is talking to whom\, not just what is said. However\, existing systems are either vulnerable to traffic analysis attacks–attacks where adversaries observe and correlate the network traffic of users–or are forced to rely on unrealistic and unenforceable assumptions about how users behave. Worse\, existing theory lacks tools to rigorously model traffic analysis attacks\, much less inform whether if a system is secure against traffic analysis or how to design systems that are. \nWe make several contributions toward our goal of practical anonymity systems that resist traffic analysis. First\, we develop the first formal framework for describing the security of systems against traffic analysis attacks\, allowing us to quantitatively describe and compare the security of all existing works. Second\, leveraging this framework\, we develop a security definition that distinguishes between systems that are and are not susceptible to traffic analysis. We call this property input/output independence. We use this definition to prove that the dominant model of systems–synchronous systems–cannot practically provide input/output independence. We then design a new asynchronous anonymity functionality\, deferred retrieval\, that achieves input/output independence with far more flexible user assumptions and up to 3400 times less traffic overhead for the same latency compared to prior methods. Finally\, we design and implement Sparta\, a family of high-throughput\, scalable instantiations of deferred retrieval using trusted execution environments and oblivious algorithms\, yielding the first practical anonymity systems that are formally resistant to long-term traffic analysis. \nEvent Host: Kyle Fredrickson\, Ph.D. Candidate\, Computer Science and Engineering \nAdvisor: Darrell Long \nZoom – https://ucsc.zoom.us/j/98133127429?pwd=QNICsMrQa6bQUKNPo40PthZyQEQCFl.1 \nPasscode – 242206
URL:https://events.ucsc.edu/event/fredrickson-k-cse-practical-anonymity-with-formal-resistance-to-traffic-analysis/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/02/ph.d.-presentation-graphic-option2.jpg
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260218T110000
DTEND;TZID=America/Los_Angeles:20260218T121500
DTSTAMP:20260428T120807
CREATED:20260210T212856Z
LAST-MODIFIED:20260210T212955Z
UID:10009195-1771412400-1771416900@events.ucsc.edu
SUMMARY:CSE Colloquium - Query Optimization: How to design a Meta-Algorithm that designs Algorithms?
DESCRIPTION:Presenter: Mahmoud Abo Khamis\, RelationalAI \nAbstract: \nDatabase systems have evolved from simple bookkeeping tools to comprehensive data analytics platforms capable of learning from the data and making business decisions. As a result\, database queries expanded in their expressive power and applications to include tensor computations\, constraint satisfaction problems\, graph analytics\, scientific computing\, SAT solving\, among others. This puts a lot of pressure on modern query optimizers to rise up to the occasion and produce efficient query plans for a wide variety of very complex queries that describe problems in different domains. The ultimate goal of query optimization is for the query optimizer to become a “meta-algorithm” where you can feed in any problem definition and get back an efficient algorithm for this particular problem. \nIn this talk\, we describe two related frameworks for query optimization that aim to take us one step in the direction of the above goal. The first framework is based on information theory. It uses information theory to get provably accurate cost estimates for query plans and to find the best query plan. Among other applications\, this framework currently achieves the best known complexity for graph pattern matching problems\, thus subsuming and generalizing known results in this area\, where\, for decades\, algorithms used to be designed by hand for specific graph patterns. The second framework is based on algebra. It uses algebraic abstractions to unify and generalize algorithms across different domains\, in the same way template programming allows for reusing code across different applications. \nBio: \nMahmoud Abo Khamis is a Senior Computer Scientist at RelationalAI\, where he has worked since 2017. He received his Ph.D. in Computer Science and Engineering from the State University of New York at Buffalo in 2016. Prior to joining RelationalAI\, he was a Senior Database Engineer at Infor from 2015 to 2017. His research interests include database systems and theory\, in-database machine learning\, query optimization and evaluation\, information theory\, and beyond worst-case analysis. His work has been recognized with two Test-of-Time Awards at ACM PODS 2025 and 2026\, three Best Paper Awards at ACM SIGMOD 2025 and ACM PODS 2022 and 2016\, three ACM SIGMOD Research Highlight Awards\, and the 2016 Best CSE Dissertation Award from SUNY Buffalo. His work has also received multiple invitations to the Journal of the ACM\, ACM STOC\, and ACM TODS. He is on the Editorial Board of ACM TODS\, and serves on the program committees of ACM PODS\, ICDT\, and ICALP among others. \nHosted by: Professor Nikos Tziavelis \nLocation: Engineering 2\, Room E2-180 (*Refreshments such as coffee\, tea\, pastries\, and fresh fruit will be provided in-person.) \nZoom: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3
URL:https://events.ucsc.edu/event/cse-colloquium-query-optimization-how-to-design-a-meta-algorithm-that-designs-algorithms/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations
ATTACH;FMTTYPE=image/png:https://events.ucsc.edu/wp-content/uploads/2026/01/BElogoWHITE.png
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260211T110000
DTEND;TZID=America/Los_Angeles:20260211T121500
DTSTAMP:20260428T120807
CREATED:20260105T205936Z
LAST-MODIFIED:20260105T205936Z
UID:10008263-1770807600-1770812100@events.ucsc.edu
SUMMARY:CSE Colloquium: Incentivized Alignment for Strategic Agents (Human and Otherwise)
DESCRIPTION:Presenter: Grant Schoenebeck\, University of Michigan \nAbstract: Advances in machine learning enable new forms of human-AI collaboration\, but collaborative settings typically involve agents with divergent objectives and private information. This will become increasingly critical in the emerging world of agentic AI\, where ML-powered agents act on behalf of individuals or institutions with conflicting goals. I use the term incentivized alignment to describe the approach of combining both machine learning and incentive design to achieve alignment of system outcomes despite misaligned agents. This talk presents two case studies of incentivized alignment showing how machine learning can make mechanism design scalable and practical\, and how mechanism design can make machine learning strategically robust. First\, I examine the use of LLMs as judges for rating subjective responses. While LLMs perform well on existing datasets\, they are highly susceptible to manipulation. I propose adapting peer-prediction mechanisms to create strategically-robust scoring mechanisms that incentivize honest reporting. Beyond ensuring high-quality inputs to AI systems\, these mechanisms can potentially eliminate reward hacking in ML training pipelines. Second\, I consider collective decision-making where agents hold different objectives and private information. The goal is to design mechanisms that incentivize strategic agents to select outcomes that would be optimal under full information sharing\, according to certain criteria. Both case studies demonstrate solutions for incentivized alignment in multi-agent systems employing the combination of incentive design and machine learning\, a theme likely to be central to the future of collaborative AI. \nBio: Grant Schoenebeck is an associate professor at the University of Michigan in the School of Information. His work has recently focused on developing and analyzing systems for eliciting and aggregating information from a diverse group of agents with varying information\, interests\, and abilities by combining ideas from machine learning and economics (e.g. game theory\, mechanism design\, and information design). More generally\, his recent work has been about incentives and (machine) learning in a variety of contexts. His research is supported by multiple NSF grants including a CAREER award and spans publications in top venues including NeurIPS\, ICLR\, EC\, WINE\, the Web Conference\, STOC\, and FOCS. His former PhD students and postdocs now hold tenure-track positions at the University of Illinois Urbana-Champaign\, Peking University\, George Mason University\, and Shanghai Jiao Tong University. He recently served as Program Committee Co-chair for WINE\, Theory Track Co-chair for EC\, and Economics and Computation Track co-chair at the Web Conference. Grant received his PhD at UC Berkeley\, studied theology at Oxford University\, and received his BA in mathematics and computer science from Harvard. \nHosted by: Professor Nikos Tziavelis \nLocation: Engineering 2\, Room E2-180 \n*Light refreshments such as coffee\, pastries\, and fruit will be available. \nZoom: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3
URL:https://events.ucsc.edu/event/cse-colloquium-incentivized-alignment-for-strategic-agents-human-and-otherwise/
LOCATION:Engineering 2\, Engineering 2 1156 High Street\, Santa Cruz\, CA\, 95064
CATEGORIES:Lectures & Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/01/portrait-scaled-e1767646696522.jpeg
GEO:37.0009723;-122.0632371
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Engineering 2 Engineering 2 1156 High Street Santa Cruz CA 95064;X-APPLE-RADIUS=500;X-TITLE=Engineering 2 1156 High Street:geo:-122.0632371,37.0009723
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260209T130000
DTEND;TZID=America/Los_Angeles:20260209T143000
DTSTAMP:20260428T120807
CREATED:20260127T195054Z
LAST-MODIFIED:20260127T195054Z
UID:10009120-1770642000-1770647400@events.ucsc.edu
SUMMARY:Li\, X. (CSE) - Compute-Efficient Scaling of Fully-Open Visual Encoders
DESCRIPTION:Vision encoders have demonstrated significant performance gains in visual generation and multimodal reasoning. These improvements are primarily attributed to the scaling of data\, model capacity\, and compute. However\, this progress is becoming less accessible due to a lack of transparency in data curation and training recipes. In combination with the high compute requirements of foundation-scale pre-training\, these factors hinder independent reproducibility. \nIn this dissertation\, we democratize large-scale visual encoder training by developing compute-efficient\, reproducible training recipes for video encoders\, vision-language models (VLMs)\, and multimodal large language models (MLLMs). First\, we challenge the common belief that scaling necessarily requires proportionally more resources. Specifically\, we show that decoupled pre-training separates key factors such as space/time and token length\, and learns strong priors first. This design yields dramatic efficiency gains across image\, video\, and generative modeling. Next\, we address the challenge of undisclosed or inaccessible training data by releasing and systematically studying the curation of high-quality\, large-scale datasets. We demonstrate that high-quality synthetic captions at scale enable vision-language models to learn stronger visual representations\, especially when paired with training frameworks that unify contrastive and generative objectives. Lastly\, building on these findings\, we develop fully open vision encoders with complete training data\, recipes\, and checkpoints\, and show that transparency can enable rather than hinder state-of-the-art performance as an MLLMs’ visual backbone. \nTogether\, these contributions establish that openness and efficiency are mutually reinforcing\, providing a reproducible foundation for the next generation of visual intelligence. \nEvent Host: Xianhang Li\, Ph.D. Candidate\, Computer Science and Engineering \nAdvisor: Cihang Xie  \nZoom- https://ucsc.zoom.us/j/95801462664?pwd=koENnyV65jyPnkJYTbiYr1jaNsV5BE.1 \nPasscode- 782017
URL:https://events.ucsc.edu/event/li-x-cse-compute-efficient-scaling-of-fully-open-visual-encoders/
CATEGORIES:Ph.D. Presentations
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/01/ph.d.-presentation-graphic-option2-1.jpg
LOCATION:
END:VEVENT
END:VCALENDAR