Home
Lectures
Temporal Graph Learning for Financial World: Algorithms, Scalability, Explainability & Fairness

Temporal Graph Learning for Financial World: Algorithms, Scalability, Explainability & Fairness

Speaker: Nitendra Rajput – Gurgaon, India
Topic(s): Artificial Intelligence, Machine Learning, Computer Vision, Natural language processing

Abstract

The most intuitive way to model a transaction in the financial world is through a Graph. Every transaction can be considered as an edge between two vertices, one of which is the paying party and another is the receiving party. Properties of these nodes and edges directly map to business problems in the financial world. The problem of detecting a fraudulent transaction can be considered as a property of the edge. The problem of money laundering can be considered as a path-detection in the Graph. The problem of a merchant going delinquent can be considered as the property of a node. While there are many such examples, the above help in realising the direct mapping of Graph properties with the financial problems in the real-world.

This talk will build upon this promise of using Graph based Learning for solving business problems of the financial world. We will start with representing transactions as a Graph. We will then get into the specifics of the transaction graph and highlight the challenge with respect to the transactions graph. These challenge are related to (a) the size of the graph (over 100 million nodes and billions of edges), (b) the temporal nature of the graph (new transactions keep appearing every second, new merchants and cards keep appearing every day - thus the graph is always evolving), and, (c) financial world being a heavily regulated industry, the questions around interpretability and fairness are critical to them being of any practical use.

After setting the business context, this talk will present latest developments in machine learning from large graphs. We will show how some of our work (BipGNN) and that in the research community (DGL) has scaled to graphs of over billion edges and 100 million nodes. We will then present the research in Temporal Graph Networks to handle dynamic graphs. Finally, we will highlight the challenges with respect to the bias in Graph based algorithms and discuss ways to mitigate it.

About this Lecture

Number of Slides: 100
Duration: 150 minutes
Languages Available: English
Last Updated:

Request this Lecture

To request this particular lecture, please complete this online form.

Request a Tour

To request a tour with this speaker, please complete this online form.

All requests will be sent to ACM headquarters for review.