BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240626T180033Z
LOCATION:3012\, 3rd Floor
DTSTART;TZID=America/Los_Angeles:20240626T164500
DTEND;TZID=America/Los_Angeles:20240626T170000
UID:dac_DAC 2024_sess154_RESEARCH1530@linklings.com
SUMMARY:PPGNN: Fast and Accurate Privacy-Preserving Graph Neural Network I
 nference via Parallel and Pipelined Arithmetic-and-Logic FHE Accelerator
DESCRIPTION:Research Manuscript\n\nYuntao Wei, Xueyan Wang, Song Bian, Yic
 heng Huang, and Weisheng Zhao (Beihang University) and Yier Jin (Universit
 y of Science and Technology in China)\n\nGraph Neural Networks (GNNs) are 
 increasingly used in fields like social media and bioinformatics, promotin
 g the prosperity of cloud-based GNN inference services. Nevertheless, data
  privacy becomes a critical issue when handling sensitive information. Ful
 ly Homomorphic Encryption (FHE) enables computations on encrypted data, wh
 ile privacy-preserving GNN inference generally necessitates ensuring graph
  structure data confidentiality and maintaining computation precision, bot
 h of which are computationally expensive in FHE. Existing schemes of GNNs 
 inference with FHE are deterred by either computational overhead, accuracy
  degradation, or incomplete data protection. This paper presents PPGNN to 
 address these challenges all at once.  We first propose a novel privacy-pr
 eserving GNN inference algorithm utilizing a high-accuracy arithmetic-and-
 logic FHE approach, meanwhile only need much smaller parameters, substanti
 ally reducing computational complexity and facilitating parallel processin
 g. Correspondingly, a dedicated hardware architecture has been designed to
  implement these innovations, with featured specialized units for arithmet
 ic and logic FHE operations in a pipelined manner. Collectively, PPGNN ach
 ieves 2.7× and 1.5× speedup over state-of-the-art Arithmetic FHE and Logic
  FHE accelerators while ensuring high accuracy, simultaneously with about 
 18× energy reduction on average.\n\nTopic: Security\n\nKeyword: Hardware S
 ecurity: Primitives, Architecture, Design & Test\n\nSession Chair: Dean Su
 llivan (University of New Hampshire)
END:VEVENT
END:VCALENDAR