BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:SIGNET SEMINAR - Zirou Qiu
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260718T091456Z
UID:tag:localist.com\,2008:EventInstance_36085569069776
DTSTART:20210303T160000Z
DTEND:20210303T171500Z
DESCRIPTION:ELRUNA: Elimination Rule-based Network Alignment\n\n \n\nAbstra
 ct: Networks model a variety of complex phenomena across different domains
 . In many applications\, one of the most essential tasks is to align two o
 r more networks to infer the similarities between cross-network vertices a
 nd discover potential node-level correspondence. In this paper\, we propos
 e ELRUNA (Elimination rule-based network alignment)\, a novel network alig
 nment algorithm that relies exclusively on the underlying graph structure.
  Under the guidance of the elimination rules that we defined\, ELRUNA comp
 utes the similarity between a pair of cross-network vertices iteratively b
 y accumulating the similarities between their selected neighbors. The resu
 lting cross-network similarity matrix is then used to infer a permutation 
 matrix that encodes the final alignment of cross-network vertices. In addi
 tion to the novel alignment algorithm\, we also improve the performance of
  local search\, a commonly used post-processing step for solving the netwo
 rk alignment problem\, by introducing a novel selection method RAWSEM (Ran
 dom walk based selection method) based on the propagation of the levels of
  mismatching (defined in the paper) of vertices across the networks. The k
 ey idea is to pass on the initial levels of mismatching of vertices throug
 hout the entire network in a random-walk fashion. Through extensive numeri
 cal experiments on real networks\, we demonstrate that ELRUNA significantl
 y outperforms the state-of-the-art alignment methods in terms of alignment
  accuracy under lower or comparable running time. Moreover\, ELRUNA is rob
 ust to network perturbations such that it can maintain a close to optimal 
 objective value under a high level of noise added to the original networks
 . Finally\, the proposed RAWSEM can further improve the alignment quality 
 with a less number of iterations compared with the naive local search meth
 od.\n\n \n\n \n\nThe paper is accepted at ACM J of Experimental Algorithmi
 cs Arxiv preprint: https://arxiv.org/abs/1911.05486   \n\n \n\nCollaborato
 rs: \n\nRuslan Shaydulin (Argonne National Lab)\n\nXiaoyuan Liu (UD)\n\nYu
 ri Alexeev (Argonne National Lab)\n\nChristopher S. Henry (Argonne Nationa
 l Lab)\n\nIlya Safro (UD)
LOCATION:
SUMMARY:SIGNET SEMINAR - Zirou Qiu
URL;VALUE=URI:https://events.udel.edu/event/signet_seminar_-_zirou_qiu
CATEGORIES:Academics
CATEGORIES:College of Engineering
END:VEVENT
END:VCALENDAR