BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Graph Neural Networks for Geometric Graphs - Chaitanya K. Joshi\, 
 Simon V. Mathis
DTSTART:20221108T130000Z
DTEND:20221108T140000Z
UID:TALK183641@talks.cam.ac.uk
CONTACT:Mateja Jamnik
DESCRIPTION:Join us in Lecture Theatre 2 or on "Zoom":https://zoom.us/j/99
 166955895?pwd=SzI0M3pMVEkvNmw3Q0dqNDVRalZvdz09\n\nGeometric graphs are spa
 tially embedded graphs used to model systems in biochemistry\, physical si
 mulations and multiagent robotics. Importantly\, graph attributes transfor
 m along with global Euclidean transformations or symmetries of the system\
 , such as rotations\, reflections\, and translation. Graph Neural Networks
  (GNNs) with global symmetries 'baked in' have emerged as the architecture
  of choice for geometric graphs. This talk will introduce two classes of G
 eometric GNNs: (1) *Equivariant GNNs*\, which use both scalar and geometri
 c features that are equivariant to global symmetries\; and (2) *Invariant 
 GNNs*\, which only reason locally via invariant scalars such as distances 
 and angles. Additionally\, we will study the expressive power of the two c
 lasses of Geometric GNNs from the perspective of distinguishing geometric 
 graphs\, i.e. graph isomorphism testing. We will introduce a *Geometric We
 isfeiler-Leman* graph isomorphism test (GWL). We will then use the GWL fra
 mework to formally show that equivariant GNNs have greater expressive powe
 r than invariant GNNs\, as they enable propagating geometric information b
 eyond local neighbourhoods and compositionally build long-range interactio
 ns.\n\nThis talk is based on the paper *"On the Expressive Power of Geomet
 ric Graph Neural Networks"*\, by Chaitanya K. Joshi (x)\, Cristian Bodnar 
 (x)\, Simon V. Mathis\, Taco Cohen\, and Pietro Liò\, to be presented as 
 an Oral at the _NeurIPS 2022 Workshop on Symmetry and Geometry in Neural R
 epresentations_.\n\n\n\n
LOCATION:Lecture Theatre 2 and Zoom
END:VEVENT
END:VCALENDAR