Skip to main content
Skip to main menu Skip to spotlight region Skip to secondary region Skip to UGA region Skip to Tertiary region Skip to Quaternary region Skip to unit footer

Slideshow

Agus Sudjianto

Agus Sudjianto
Agus Sudjianto
Executive Vice President, Head of Corporate Model Risk
Wells Fargo & Company
Room 204, Caldwell Building
Event Flyer (175.39 KB)

Machine Learning for High-Risk Applications in Banking

Renowned statistician George Box once famously stated, “All models are wrong, but some are useful.” In a world where machine learning increasingly automates important decisions about our lives, the consequences of model failures can be catastrophic. It’s critical to take deliberate steps to mitigate risk and avoid unintended harm.



Evaluating the conceptual soundness of machine learning models, which includes explainability and interpretability, can be challenging due to their over-parameterization. There have been many attempts in the field of so-called Explainable Artificial Intelligence (XAI), which are typically done by employing auxiliary tools, so-called post-hoc explainers, to explain black box models. Though useful, these tools have their weaknesses due to their approximation nature and are not adequate for high-risk applications. I will discuss how we design interpretability for complex machine learning models, including the use of Functional ANOVA as a constraint to impose interpretability.



Rigorous and comprehensive model testing is a key requirement for high-risk applications, and standard model performance testing is insufficient as models in the real world live in non-IID and changing environments. Model testing must be able to uncover model weaknesses, including their robustness against input corruption, their ability to provide reliable predictions, and their performance resilience under distribution shifts. In this talk, I will present our practice today and the technical challenges we face.

 

Lunch Talk: 12:00pm

Room 434, Cohen Room, Brooks Hall

Being a Quant/ Data Scientist in Banking

Abstract:

This presentation will discuss the role of Quant/Data Scientist in banking. I will discuss various applications of modeling in banking and how the job is shaped by the adoption of machine learning. I will discuss what skills and behaviors that students can prepare to have successful careers. This will be an informal conversation about possible career paths with time for interactions and questions.

Biography

Dr. Agus Sudjianto is an executive vice president, head of Model Risk and a member of Management Committee at Wells Fargo, where he is responsible for enterprise model risk management.

Prior to his current position, Agus was the modeling and analytics director and chief model risk officer at Lloyds Banking Group in the United Kingdom. Before joining Lloyds, he was an executive and head of Quantitative Risk at

Bank of America.

Prior to his career in banking, he was a product design manager in the Powertrain Division of Ford Motor Company.

Agus holds several U.S. patents in both finance and engineering. He has published numerous technical papers and is a co-author of Design and Modeling for Computer Experiments. His technical expertise and interests

include quantitative risk, particularly credit risk modeling, machine learning and computational statistics.

He holds masters and doctorate degrees in engineering and management from Wayne State University and the Massachusetts Institute of Technology.

Support us

We appreciate your financial support. Your gift is important to us and helps support critical opportunities for students and faculty alike, including lectures, travel support, and any number of educational events that augment the classroom experience. Click here to learn more about giving.

Every dollar given has a direct impact upon our students and faculty.