Intro to Graph Convolutional Networks(GCN)

Apr 24, 2018 · Austin, United States of America

Speaker:Graham Ganssle, Ph.D., P.G.
Head of Data Science, Expero Inc

Topic: Intro to Graph Convolutional Networks(GCN)

Abstract:
Is this group of delis a money laundering ring? or are they simply exchanging provolone?Why does Devin have so many Facebook friends, and I only have a handful?
The architecture of the algorithm described in this presentation is based on the work of Kipf and Welling on GCN and The database represented in this work is Neo4j.

KipF and Welling Paper:
https://arxiv.org/abs/1609.02907

"The answer to one of these questions is obvious (because I’m a nerd giving an ML presentation), but both can be solved with graph convolutional networks. GCNs use a novel neural network kernelization which generalizes the accuracy and speed of CNNs to non-Euclidean information spaces. Join me for a chat about the applications of GCNs, the elegance of convolutional mathematics, and what the hell non-Euclidean means. I’ll even send you home with the party favor of GCN open source software." - Graham

Speaker Bio:
Graham is Head of Data Science at Expero Inc. He is a recovering physicist, deep learning his way through a world of digital signals. He is the head of data science at Expero Inc, where, with the synergy of his expert team, he uses data to explore the domains of finance, national security, energy, and health care.

Agenda
6:30 - 7:00: Registration, Pizza & Soft drinks
7:00 - 7:30: Presentation on topic
7:30 - 7:45: Demonstration of topic
7:45 - 8:15: Lightning Talk + Networking

Event organizers

Are you organizing Intro to Graph Convolutional Networks(GCN) ?

Claim the event and start manage its content.

I am the organizer
Social
Rating

based on 0 reviews