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:20260108T110000
DTEND;TZID=Asia/Jerusalem:20260108T120000
DTSTAMP:20260417T140526
CREATED:20260105T064430Z
LAST-MODIFIED:20260105T064430Z
UID:5077-1767870000-1767873600@cis.haifa.ac.il
SUMMARY:Jonathan Shafer (MIT) - From Learning Theory to Cryptography: Provable Guarantees for AI
DESCRIPTION:Title: From Learning Theory to Cryptography: Provable Guarantees for AI \nAbstract: \nEnsuring that AI systems behave as intended is a central challenge in contemporary AI. This talk offers an exposition of provable mathematical guarantees for learning and security in AI systems. \nStarting with a classic learning-theoretic perspective on generalization guarantees\, we present two results quantifying the amount of training data that is provably necessary and sufficient for learning: (1) In online learning\, we show that access to unlabeled data can reduce the number of prediction mistakes quadratically\, but no more than quadratically [NeurIPS23\, NeurIPS25 Best Paper Runner-Up]. (2) In statistical learning\, we discuss how much labeled data is actually necessary for learning—resolving a long-standing gap left open by the celebrated VC theorem [COLT23]. \nProvable guarantees are especially valuable in settings that require security in the face of malicious adversaries. The main part of the talk adopts a cryptographic perspective\,  showing how to: (1) Utilize interactive proof systems to delegate data collection and AI training tasks to an untrusted party [ITCS21\, COLT23\, NeurIPS25]. (2) Leverage random self-reducibility to provably remove backdoors from AI models\, even when those backdoors are themselves provably undetectable [STOC25]. \nThe talk concludes with an exploration of future directions concerning generalization in generative models\, and AI alignment against malicious and deceptive AI.
URL:https://cis.haifa.ac.il/event/jonathan-shafer-mit-from-learning-theory-to-cryptography-provable-guarantees-for-ai/
CATEGORIES:סמינרים מדעי המחשב
END:VEVENT
END:VCALENDAR