Abstract:Fairness and accountability are cornerstones in regulated industries such as finance and insurance. Explainability is expected of all machine learning models in order to comply with strict
Fairness and accountability are cornerstones in regulated industries such as finance and insurance. Explainability is expected of all machine learning models in order to comply with strict audit trails and regulatory oversight. It is often challenging to bridge the gap between the realm of machine learning and regulatory needs.
This session is for machine learning scientists working in regulated industries to learn how to structure and present their work in a way that meets the needs of regulated industries. The session will describe a case study of how a machine learning team worked with an actuarial team to test models for health and life insurance underwriting and ensure sufficient explainability was achieved. The case study will offer a framework to help machine learning teams bridge the gap and get their models into production.
What You’ll Learn:
There are currently no right answers or standards on the expectations of explainability required for machine learning models in regulated industries. The session will provide a case study of how our team went through the process and learnings form our trial and errors to get a better understanding of what’s expected.
Spark Tseung is an Applied Data Scientist at Knowtions Research where he focuses on building frameworks for actuarial and underwriting validation to help insurers use machine learning to protect more people. Spark is working towards his PhD in Statistics and specializes in the application of machine learning methods in Property & Casualty loss modelling and risk selection.
Hanieh Arjmand is a Machine Learning Researcher at Knowtions Research where she focuses on discovering and applying the best machine learning techniques to healthcare and insurance problems to help insurers use machine learning to protect more people. Hanieh completed her MSc (PhD candidate) in Biomedical Engineering and specializes in Machine Learning, Deep Learning, Image Analysis and Computer Vision.
(Thursday) 10:55 AM - 11:40 AM