Mon 11 Apr 2022 - Deep neural network models have substantial advantages over traditional and machine learning methods that make this class of models, particularly promising for adoption by actuaries. Nonetheless, several important aspects of these models have not yet been studied in detail in the actuarial literature: the effect of hyperparameter choice on the accuracy and stability of network predictions, methods for producing uncertainty estimates, and the design of deep learning models for explainability. To allow actuaries to incorporate deep learning safely into their toolkits, we review these areas in the context of a deep neural network for forecasting mortality rates.
Greg Solomon works as an independent consulting actuary out of Hong Kong, where he has been living for the last decade. Under his Eigengrey brand, Greg also does advisory work in the fintech and insurtech space. Prior to this, Greg was with Swiss Re for over 20 years in 4 countries, he ran the global Peak Re L&H business, and set up and grew the Willis Towers Watson Asian life reinsurance broking unit. Greg believes in actuaries and in deep learning models, but has been waiting for someone like Ronald Richman to expertly connect them together.
Chief Actuary at Old Mutual Insure
Ron Richman is Chief Actuary at Old Mutual Insure, where he is responsible for oversight of all actuarial activities in the OMI Group. Ron is a Fellow of the IFoA and the Actuarial Society of South Africa. He holds practicing certificates in short term insurance and life insurance from ASSA. Ron also has a Masters of Philosophy in Actuarial Science, with distinction, from the University of Cape Town.
He chairs the Actuarial Society of South Africa’s climate change committee and is the vice chair of the ASTIN board. Ron has a keen professional interest in the application of AI techniques to actuarial work and has published several papers demonstrating how this can be done.