BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Events - ECPv6.18.0//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
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:20260520T110000
DTEND;TZID=America/Los_Angeles:20260520T121500
DTSTAMP:20260518T155149Z
CREATED:20260518T155149Z
LAST-MODIFIED:20260518T155149Z
UID:10014650-1779274800-1779279300@events.ucsc.edu
SUMMARY:CSE Colloquium - Safety Alignment of LMs via Non-cooperative Games
DESCRIPTION:Presenter: Arman Zharmagambetov\, Meta \nAbstract:\nEnsuring the safety of language models (LMs) while maintaining their usefulness remains a critical challenge in AI alignment. Current approaches rely on sequential adversarial training: generating adversarial (harmful) prompts and fine-tuning LMs to defend against them. We introduce a different paradigm: framing safety alignment as a non-zero-sum game between an Attacker LM and a Defender LM trained jointly via online reinforcement learning. Each LM continuously adapts to the other’s evolving strategies\, driving iterative improvement. Our method uses a preference-based reward signal derived from pairwise comparisons instead of point-wise scores\, providing more robust supervision and potentially reducing reward hacking. Our RL recipe\, AdvGame\, shifts the Pareto frontier of safety and utility\, yielding a Defender LM that is simultaneously more helpful and more resilient to adversarial attacks. In addition\, the resulting Attacker LM converges into a strong\, general-purpose red-teaming agent that can be directly deployed to probe arbitrary target models. \nBio:\nArman Zharmagambetov is a research scientist in the Fundamental AI Research (FAIR) team at Meta. His research primarily focuses on machine learning and optimization\, recently exploring their application in enhancing the security and robustness of AI systems. He received his PhD from the University of California – Merced. Afterward\, he completed his postdoctoral research at FAIR\, focusing on Reinforcement Learning\, AI-guided design and Optimization. \nHosted by: Professor Alvaro Cardenas and Professor Sungjin Im \nDate and Time: Wednesday\, May 20\, 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.) \nZoom Option: https://ucsc.zoom.us/j/93445911992?pwd=YkJ2TQtF79h0PcNXbEcpZLbpK0coiY.1&jst=3
URL:https://events.ucsc.edu/event/cse-colloquium-safety-alignment-of-lms-via-non-cooperative-games/
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:20260520T132000
DTEND;TZID=America/Los_Angeles:20260520T142500
DTSTAMP:20260513T212109Z
CREATED:20260513T211954Z
LAST-MODIFIED:20260513T212109Z
UID:10014632-1779283200-1779287100@events.ucsc.edu
SUMMARY:BME80G Seminar – Ann Mc Cartney\, "The Why\, What and How of Indigenous Data Sovereignty"
DESCRIPTION:Presenter: Dr. Ann Mc Cartney\n\n\nLocation: Virtual. Please register here: https://ucsc.zoom.us/meeting/register/ciShTZsyRViYxMDjCc_cAQ#/registration \nAbstract: In 2007 the General Assembly of the United Nations adopted the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP) that supports Indigenous Peoples’ inherent rights to self-determination and governance over Indigenous Peoples\, territories and resources. This codification in an international treaty led to the conceptualization of Indigenous Data Sovereignty in 2016. Indigenous Data Sovereignty is defined as the right of Indigenous Peoples to own\, control\, access and possess data that derive from them\, and which pertain to Nation membership\, knowledge systems\, customs or territories and is not an internationally recognized concept. Practically\, Indigenous Data Sovereignty gives decision-making authority to Indigenous Peoples and allows them to decide how best data should be collected\, accessed\, and used for nation-building. For this reason\, Indigenous Data Sovereignty is inextricably linked to Indigenous Data Governance. \nThis lecture will dive into the why\, what\, and how of Indigenous Data Sovereignty. Beginning with dive into the history of genomics research and Indigenous Peoples. Then moving into a more in-depth description of both Indigenous Data Sovereignty and Indigenous Data Governance. The lecture will close with highlighting three real world case-studies of genomics projects operationalising Indigenous Data Sovereignty. \nAbout the speaker: Dr. Mc Cartney is a Full Researcher at the University of California\, Irvine and a Consultant for the Secretariat of the Convention on Biological Diversity\, where she works at the intersection of bioinformatics\, bioethics and policy across human and non-human species. Dr. Mc Cartney completed her PhD in Bioinformatics and Molecular Evolution from Dublin City University. After this she carried out a Genomics Aotearoa Postdoctoral Research Fellowship where she created culturally respectful genomic informatics pipelines for the project that were responsive to the needs and rights of Māori. She then moved Stateside and carried out a second Visiting Postdoctoral Fellowship in the Genome Informatics Section at the NHGRI. During this time\, she also conducted a detail in the NIH Office of Science Policy. Dr. Mc Cartney then transitioned to the University of California\, Santa Cruz’s Genomics Institute as an Assistant Researcher where she worked at the forefront of genomic sequencing technologies to develop more just genomic data infrastructures\, policies and frameworks for engagement. She became an elected member of the Executive Council of the Earth BioGenome Project\, the Vice-chair of the European Reference Genome Atlas\, and the Director of International Partnerships for the Human Pangenome Reference Consortium. She also acted as a “technical expert” for many international science policy initiatives including the DSI Scientific Network\, Native Biodata Consortium and the Informal Advisory Group on DSI for the Convention on Biological Diversity. Additionally\, Dr. Mc Cartney was part of the team to build the first NIH-funded Tribal Data Repository for American Indian and Alaska Natives.
URL:https://events.ucsc.edu/event/bme80g-seminar-ann-mc-cartney-the-why-what-and-how-of-indigenous-data-sovereignty/
LOCATION:https://events.ucsc.edu/event/bme80g-seminar-ann-mc-cartney-the-why-what-and-how-of-indigenous-data-sovereignty/
CATEGORIES:Lectures & Presentations,Seminars
ATTACH;FMTTYPE=image/jpeg:https://events.ucsc.edu/wp-content/uploads/2026/05/6-6-23-Ann-Mccartney-CL-006-scaled.jpg
END:VEVENT
END:VCALENDAR