A graph, or network, is a structure that encodes pairwise relationships and a graph signal is a function defined on the nodes of the graph. The values of the weights on the edges of the graph encode an expectation on the relationship between the respective signal components. A large weight indicates that we expect the signal elements to be similar and a small weight indicates no such expectation except for what is implied by their common proximity to other nodes. The goal of graph signal processing (GSP) is to generalize the classical signal processing toolbox to graph signals.

Graph signal processing applications arise whenever we encounter one of the many signals that are supported on a graph. Examples of applications that will be showcased in the workshop include gene expression patterns defined on top of gene networks, the spread of epidemics over a social network, the congestion level at the nodes of a telecommunication network, and patterns of brain activity defined on top of a brain network.

Besides particular applications, the workshop will also showcase the advancement of the understanding of network data by redesigning traditional tools originally conceived to study signals defined on regular domains (such as time-varying signals or spatially varying images and fields) and extending them to analyze signals on the more complex graph domain. Examples of topics to be showcased in this theoretical track include graph transforms, sampling theorems, and filter design.

The graph signal processing workshop took place **June 6-8th 2018 in EPFL**

### Organizing committee

**Chair:**- Pascal Frossard, EPFL

**Technical program:**- Pierre Borgnat, ENS Lyon
- Xiaowen Dong, University of Oxford

**Invited speakers:**- Dorina Thanou, SDSC
- Nicolas Tremblay, CNRS

**Local Arrangements:**- Mireille El Gheche, EPFL
- Bastien Pasdeloup, EPFL

**Finances:**- Dorina Thanou, SDSC

**Publicity:**- Thomas Maugey, INRIA

**Web:**- Nicolas Aspert, EPFL

### Steering committee

- Pascal Frossard
- Jelena Kovacevic
- Jose Moura
- Antonio Ortega
- Mike Rabbat
- Alejandro Ribeiro