Graph Neural Networks – from a complicated data structure to meaningful insights
Graph Neural Networks – from a complicated data structure to meaningful insights
Zohar is a Data Scientist at PayPal’s horizontal Data Science team, which develops ML infrastructures that serve data scientists globally. She currently researches graph and linking solutions that serve ML solutions across various business domains at PayPal, like fraud detection, credit risk, marketing, and many more. Zohar holds a B.Sc. in Industrial Engineering from Tel Aviv University
Zohar is a Data Scientist at PayPal’s horizontal Data Science team, which develops ML infrastructures that serve data scientists globally. She currently researches graph and linking solutions that serve ML solutions across various business domains at PayPal, like fraud detection, credit risk, marketing, and many more. Zohar holds a B.Sc. in Industrial Engineering from Tel Aviv University
Graphs are an excellent method to express complicated relationships and interdependencies between objects. It is widely used in traffic problems, malicious networks detection, social networks, biomedical networks, and more. Nonetheless, their complex structure is uniquely challenging when trying to extract insights in a way that would genuinely complement traditional ML techniques. At PayPal, we are dealing with massive scale graphs. We use advanced graph embedding methods based on Graph Neural Networks and apply them in a distributed manner to our production environment. In this poster, I will describe the challenges of graph processing as a non-euclidean data object and how to graph embedding can help overcome them. I will share how we used DGL (Deep Graph Library) to implement one of the most breakthrough graph embedding algorithms (GraphSAGE) for different applications and provide unique incremental insights to our models
Graphs are an excellent method to express complicated relationships and interdependencies between objects. It is widely used in traffic problems, malicious networks detection, social networks, biomedical networks, and more. Nonetheless, their complex structure is uniquely challenging when trying to extract insights in a way that would genuinely complement traditional ML techniques. At PayPal, we are dealing with massive scale graphs. We use advanced graph embedding methods based on Graph Neural Networks and apply them in a distributed manner to our production environment. In this poster, I will describe the challenges of graph processing as a non-euclidean data object and how to graph embedding can help overcome them. I will share how we used DGL (Deep Graph Library) to implement one of the most breakthrough graph embedding algorithms (GraphSAGE) for different applications and provide unique incremental insights to our models
| 8:45 | Reception |
|---|---|
| 9:30 | Opening words by WiDS TLV ambassadors Or Basson and Noah Eyal Altman |
| 9:40 | Dr. Kira Radinski - Learning to predict the future of healthcare |
| 10:10 | Prof. Yonina Eldar - Model-Based Deep Learning: Applications to Imaging and Communications |
| 10:40 | Break |
| 10:50 | Lightning talks |
| 12:20 | Lunch & Poster session |
|---|---|
| 13:20 | Roundtable session & Poster session |
| 14:05 | Roundtable closure |
| 14:20 | Break |
| 14:30 | Dr. Anna Levant - 3D Metrology: Seeing the Unseen |
| 15:00 | Aviv Ben-Arie - Counterfactual Explanations: The Future of Explainable AI? |
| 15:30 | Closing remarks |
| 15:40 | End |
© 2018-2022 WiDS TLV – Intuit. All rights reserved.
Scotty – By Nir Azoulay
Design: Sharon Geva