BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:America/New_York
X-WR-TIMEZONE:America/New_York
BEGIN:VEVENT
UID:0-7317@eng.ufl.edu
DTSTART;TZID=America/New_York:20250213T120000
DTEND;TZID=America/New_York:20250213T130000
DTSTAMP:20251201T210652Z
URL:https://www.eng.ufl.edu/news-events/events/cise-faculty-seminar-dr-sum
 it-jha/
SUMMARY:CISE Faculty Seminar: Dr. Sumit Jha
DESCRIPTION:Zoom Link: https://ufl.zoom.us/j/94143185671\nBiography: Dr. Su
 mit Kumar Jha is Eminent Scholar Chair Professor of Computer Science at Fl
 orida International University. He earned his Ph.D. in Computer Science fr
 om Carnegie Mellon University and has held multiple summer faculty appoint
 ments with the Air Force Research Laboratory Information Directorate.\nDr.
  Jha currently serves as the lead PI on projects in excess of $11 million.
  Dr. Jha has led interdisciplinary\, multi-institutional teams on projects
  funded by the National Science Foundation (NSF)\, Defense Advanced Resear
 ch Projects Agency (DARPA)\, Department of Energy (DOE)\, and other federa
 l agencies. His work has been published at premier venues\, such as AAAI\,
  DAC\, DATE\, ICCAD\, ICLR\, IJCAI\, and NeurIPS. His research focuses on 
 building high-assurance and efficient AI at both algorithmic and hardware 
 levels. He has developed methods to make AI systems more transparent and r
 esistant to threats by combining symbolic decision procedures\, human expe
 rtise\, foundation models\, and deductive reasoning. Dr. Jha has also pion
 eered flow-based in-memory computing techniques for data-intensive applica
 tions\, advancing a new paradigm for efficient and sustainable AI in his r
 esearch funded by the NSF over the past 11 years.\nDr. Jha has been a PI o
 n projects from DARPA GARD\, DARPA ANSR\, DARPA TIAMAT\, NSF Software and 
 Hardware Foundations (SHF)\, NSF Exploiting Parallelism and Scalability (X
 PS)\, NSF Scalable Parallelism in the Extreme (SPX)\, and NSF Formal Metho
 ds in the Field (FMitF)\, NGA Boosting Innovative GEOINT\, NNSA/ORNL\, ONR
  Science of AI\, Department of Energy\, AFRL\, and the Royal Bank of Canad
 a Innovation Lab. His work has earned multiple best paper awards and nomin
 ations at various forums (IEEE DATE\, ACM/IEEE ICCAD\, IEEE MILCOM\, IEEE 
 ICCABS)\, as well as the prestigious Air Force Office of Scientific Resear
 ch Young Investigator Program (AFOSR YIP) Award. Dr. Jha aims to advance h
 igh-assurance and efficient AI systems that drive innovation across scienc
 e\, engineering\, healthcare\, and sustainable peace.\nTitle of the Talk: 
 Formal Methods for High-assurance and Efficient AI\nAbstract: Deployment o
 f Artificial Intelligence (AI) in high-assurance settings demands formal g
 uarantees\, which remain challenging for the current generation of AI mode
 ls\, particularly deep neural networks. The limited scalability of the tra
 ditional formal verification methods has been further exacerbated by moder
 n foundation models\, such as large language models (LLMs). In this talk\,
  we present scalable approaches to  (1) explain the decisions of AI 
 agents in a human-interpretable manner drawing upon neural stochastic diff
 erential equations and path integrals\, (2) communicate our ethics\, d
 omain knowledge\, and feedback to AI agents informally and formally levera
 ging probabilistic temporal logics as formal specifications\, and (3) enab
 le human-in-the-loop auditing of the behavior of teams of LLM agents a
 gainst regulatory and ethical guidelines through model checking\, theorem 
 proving and provably correct symbolic AI methods. Additionally\, I will br
 iefly discuss recent progress using Binary decision diagrams and other for
 mal methods towards in-memory computing\, demonstrating how algorithm-ha
 rdware co-design can reduce the soaring energy demands of foundation model
 s. These innovations offer a blueprint for creating the next generation of
  AI that seamlessly fuses human expertise\, ethical guidelines\, domain ru
 les\, and data-driven intelligence while maintaining interpretability and 
 efficiency. 
CATEGORIES:Seminars
LOCATION:Malachowsky Hall 5210\, 1889 Museum Rd\, Gainesville\, FL\, 32611\
 , United States
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=1889 Museum Rd\, Gainesvill
 e\, FL\, 32611\, United States;X-APPLE-RADIUS=100;X-TITLE=Malachowsky Hall
  5210:geo:0,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:STANDARD
DTSTART:20241103T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
END:STANDARD
END:VTIMEZONE
END:VCALENDAR