BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Graph Joinings\, Reversible Markov Chains\, and Graph Isomorphism
DTSTART:20260313T151500
DTEND:20260313T161500
DTSTAMP:20260405T153658Z
UID:ebcb57e4c0b9f4f907097e9ce3ab4ef3ff064b0191db449c2808d226
CATEGORIES:Conferences - Seminars
DESCRIPTION:Andrew Nobel\, Chapel Hill\nThe correspondence between weighte
 d undirected graphs reversible Markov chains is elementary and well known.
   I will describe recent work that leverages this correspondence\, in con
 junction with ideas from ergodic theory\, to study the structural discorda
 nce of graphs and Markov chains via graph joinings.  Informally\, a joini
 ng of two graphs is a graph on the product of their vertex sets giving ris
 e to a coupling of their random walks.\nTwo graphs are strongly disjoint i
 f their only joining is the tensor product\, and are weakly disjoint if th
 e degree function of every joining is equal to the degree function of the 
 tensor product. I will present spectral characterizations of strong and we
 ak disjointness\, and describe corresponding results for reversible Markov
  chains. In a different direction\, I will show how optimal joinings based
  on a vertex-label based cost function can detect and identify graph isomo
 rphisms for suitable families of graphs.\n\nJoint work with Yang Xiang\, P
 huong Hoang\, Bongsoo Yi\, and Kevin McGoff.
LOCATION:CM 1 517 https://plan.epfl.ch/?room==CM%201%20517
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR