Presented By O’Reilly and Cloudera
Make Data Work
21–22 May 2018: Training
22–24 May 2018: Tutorials & Conference
London, UK

Risk-sharing pools: Winning zero-sum games through machine learning

Baiju Devani (Aviva Canada), Etienne Chasse St-Laurent (Aviva Canada)
12:0512:45 Wednesday, 23 May 2018
Data science and machine learning
Location: Capital Suite 10/11 Level: Intermediate
Secondary topics:  Financial Services
Average rating: ***..
(3.33, 3 ratings)

Who is this presentation for?

  • Data scientists, actuaries, and business executives

Prerequisite knowledge

  • An intermediate understanding of machine learning

What you'll learn

  • Understand the limits of classical models in the insurance industry and the problems around risk-sharing pools
  • Explore an applied machine learning approach that leverages methods such as elastic nets and gradient boosted trees


Risk-sharing pools allow insurers to get rid of risks they are forced to insure in highly regulated markets. Insurers thus cede both the risk and its premium. Since all participants in the system are trying to cede their worst possible risks, this becomes a zero-sum game in which more sophisticated models win at the expense of less sophisticated models.

Traditionally, the insurance industry relies on generalized linear models for such problems. Baiju Devani and Étienne Chassé St-Laurent discuss traditional approaches and challenges and share an applied machine learning approach that leverages an ensemble of models, including elastic nets and XGBoost, to gain a distinctive market advantage. Baiju and Étienne conclude by showcasing results that not only outperform traditional methods but enable insurance companies to add tens of millions of dollars to their bottom lines.

Photo of Baiju Devani

Baiju Devani

Aviva Canada

Baiju Devani is vice president of enterprise analytics at Aviva Canada, where he leads a team of data scientists in application of analytics across all aspects of the insurance business, from core insurance activities, such as pricing and risk selection, to the development of cutting-edge robotic processes and the application of machine learning algorithms to areas such as claims processing and optimizing client acquisition. Previously, Baiju led the Analytics Group at IIROC, an entity overseeing Canadian equity and fixed-income markets, where he guided decision making in todays machine-driven (algorithmic) markets, embedded machine learning, and other algorithmic surveillance capabilities for market oversight and spearheaded IIROC’s primary research into market-structure issues and emerging technologies, such as blockchains; was part of the early team at fintech startup OANDA, which disrupted retail foreign-exchange markets, where he was responsible for building and leading the data engineering, data science, and growth teams; and founded Fstream, a SaaS provider for ingesting and analyzing high-frequency streaming data. Baiju holds both a BSc and MSc in computer science from Queen’s University. He developed his data chops working on large biological datasets as part of his graduate work and later at the Ontario Cancer Institute.

Photo of Etienne Chasse St-Laurent

Etienne Chasse St-Laurent

Aviva Canada

Étienne Chassé St-Laurent is a data scientist working in enterprise analytics at Aviva Canada, where he is building a new generation of predictive models by leveraging his background in statistics and actuarial science together with machine learning to solve issues specific to the insurance industry. In the past, he’s worked at Statistics Canada and in the pharmaceutical industry but has found a home for the last 10 years in insurance, six of them in R&D. He holds a BSc in actuarial science and an MSc in statistics. Étienne is mainly fueled by sugar and always receptive to book recommendations.