BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//הפקולטה למדעי המחשב והמידע - ECPv6.15.15//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://cis.haifa.ac.il
X-WR-CALDESC:Events for הפקולטה למדעי המחשב והמידע
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Jerusalem
BEGIN:DAYLIGHT
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:IDT
DTSTART:20250328T000000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
TZNAME:IST
DTSTART:20251025T230000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:IDT
DTSTART:20260327T000000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
TZNAME:IST
DTSTART:20261024T230000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:IDT
DTSTART:20270326T000000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
TZNAME:IST
DTSTART:20271030T230000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Jerusalem:20260129T110000
DTEND;TZID=Asia/Jerusalem:20260129T120000
DTSTAMP:20260417T144642
CREATED:20260127T063054Z
LAST-MODIFIED:20260127T063054Z
UID:5963-1769684400-1769688000@cis.haifa.ac.il
SUMMARY:Noam Razin (Princeton University) - Fundamentals of Aligning General-Purpose AI
DESCRIPTION:Abstract:\nThe field of artificial intelligence (AI) is undergoing a paradigm shift\, moving from neural\nnetworks trained for narrowly defined tasks (e.g.\, image classification and machine translation)\nto general-purpose models such as ChatGPT. These models are trained at unprecedented\nscales to perform a wide range of tasks\, from providing travel recommendations to solving\nOlympiad-level math problems. As they are increasingly adopted in society\, a central challenge\nis to ensure the alignment of general-purpose models with human preferences. In this talk\, I will\npresent a series of works that reveal fundamental pitfalls in existing alignment methods. In\nparticular\, I will show that they can: (1) suffer from a flat objective landscape that hinders\noptimization\, and (2) fail to reliably increase the likelihood of generating preferred outputs\,\nsometimes even causing the model to generate outputs with an opposite meaning. Beyond\ncharacterizing these pitfalls\, our theory provides quantitative measures for identifying when they\noccur\, suggests preventative guidelines\, and has led to the development of new data selection\nand alignment algorithms\, validated at large scale in real-world settings. Our contributions\naddress both efficiency challenges and safety risks that may arise in the alignment process. I\nwill conclude with an outlook on future directions\, toward building a practical theory in the age of\ngeneral-purpose AI. \nShort bio:\nNoam Razin is a Postdoctoral Fellow at Princeton Language and Intelligence\, Princeton\nUniversity. His research focuses on the fundamentals of artificial intelligence (AI). By combining mathematical analyses with systematic experimentation\, he aims to develop theories that shed light on how modern AI works\, identify potential failures\, and yield principled methods for improving efficiency\, reliability\, and performance.\nNoam earned his PhD in Computer Science at Tel Aviv University\, where he was advised by Nadav Cohen. Prior to that\, he obtained a BSc in Computer Science (summa cum laude) at The\nHebrew University of Jerusalem under the Amirim honors program. For his research\, Noam received several honors and awards\, including the Zuckerman Postdoctoral Scholarship\, the Israeli Council for Higher Education (VATAT) Postdoctoral Scholarship\, the Apple Scholars in AI/ML PhD fellowship\, the Tel Aviv University Center for AI and Data Science excellence fellowship\, and the Deutsch Prize for PhD candidates.
URL:https://cis.haifa.ac.il/event/noam-razin-princeton-university-fundamentals-of-aligning-general-purpose-ai/
CATEGORIES:סמינרים מדעי המחשב
END:VEVENT
END:VCALENDAR