St. Louis Machine Learning & Data Science

Genetic Algorithms - what they are and how to apply them to solve problems

September 17, 2014 · 6:30 PM
S. Taylor Avenue Building

Genetic algorithms are being used in many scientific professions to solve problems too difficult to solve with standard deterministic techniques. A genetic algorithm is an inductive, rather than deductive approach that learns as it goes along. This session will give an in-depth look at genetic algorithms - what they are, why they are important for actuaries, when to use them, and how to create them. It starts with the origin of genetic algorithms and their relation to genetics. Next, it shows a couple of applications that teach the basic principles. Later, it shows an actuarial application where the genetic algorithm approach gives a higher degree of optimization than that achieved by more traditional techniques. Finally, it provides a set of guidelines for when and how to utilize them. At the conclusion of the session, attendees will be able to: apply simple genetic algorithms to real-world problems; determine if a problem is suitable for the application of a genetic algorithm; develop further applications of genetic algorithms based on their own experiments and research; and explain genetic algorithms at a basic level to their colleagues.

Dave Snell, ASA, MAAA, FLMI, CLU, ChFC, ARA, ACS, MCP
Technology Evangelist
RGA Reinsurance Company
Dave Snell’s title is Technology Evangelist, and despite the whimsical sound, it seems to fit him. In September, 2007, Dave took early retirement from his position as VP, Asia-Pacific Technology for RGA Reinsurance Company, where, based in Sydney, Australia, he managed new and existing technology for all of Asia and Australia. That was on a Friday. On Monday, he returned home to the U.S. as a consultant to the Vice Chair of RGA, where he networks with kindred spirits among actuaries and technology associates to identify and overcome actuarial obstacles through better use of technology tools. Dave has written thousands of programs in dozens of programming languages - including an artificial intelligence based expert system for underwriting in use in over a dozen countries and several languages. A frequent speaker at Society of Actuaries (SOA) meetings, Dave is a past Chair of the SOA Actuary of the Future section, Editor of the SOA Forecasting and Futurism section newsletter, a past Chair of the SOA Technology section and a past Chair of the St. Louis Actuaries Club. He also serves on the Actuarial Science program Advisory Board at Maryville University, in St. Louis, Missouri. In what seems now like previous lifetimes, he was a Mechanical Engineer, and later a life insurance agent and brokerage manager.