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:3002\, 3rd Floor
DTSTART;TZID=America/Los_Angeles:20240627T144500
DTEND;TZID=America/Los_Angeles:20240627T150000
UID:dac_DAC 2024_sess104_RESEARCH1526@linklings.com
SUMMARY:TrafficHD: Efficient Hyperdimensional Computing for Real-Time Netw
 ork Traffic Analytics
DESCRIPTION:Research Manuscript\n\nHaodong Lu, Zhiyuan Ma, Xinran Li, Shiy
 an Bi, Xiaoming He, and Kun Wang (Fudan University)\n\nWith the evolution 
 of network infrastructure, the pattern of network traffic becomes unpreced
 entedly complex. Conventional machine learning algorithms are struggling t
 o cope with the high-dimensional data and real-time processing speeds requ
 ired in such complex networks. Fortunately, hyperdimensional Computing (HD
 C), which is power-efficient and supports parallel processing, provides a 
 potential solution to this challenge. In this paper, we present TrafficHD,
  a novel classification framework that leverages HDC to analyze network tr
 affic in real-time. By transforming network traffic features into high-dim
 ensional binary vectors, TrafficHD enables the rapid execution of recognit
 ion tasks within the constraints of real-time systems. Extensive evaluatio
 ns on a wide range of network tasks show that TrafficHD achieves 30.57× an
 d 98.32× faster than state-of-the-art (SOTA) machine learning and HDC algo
 rithms while providing 3× higher robustness to network noise.\n\nTopic: AI
 \n\nKeyword: AI/ML Application and Infrastructure\n\nSession Chairs: Cong 
 (Callie) Hao (Georgia Institute of Technology) and Haoxing “Mark” Ren (NVI
 DIA)
END:VEVENT
END:VCALENDAR