Details
- Time: 11:00-12:00AM
- Date: Friday, December 15, 2023
- Venue: MB341
- Speaker: Prof. Zhuo Jin, Macquarie University
- Host: Prof. Hailiang Yang
Abstract
This paper develops a hybrid deep reinforcement learning approach to manage an insurance portfolio for diffusion models. To address the model uncertainty, we adopt the recently developed modelling of exploration and exploitation strategies in a continuous-time decision-making process with reinforcement learning. We consider an insurance portfolio management problem in which an entropy-regularized reward function and corresponding relaxed stochastic controls are formulated. To obtain the optimal relaxed stochastic controls, we develop a Markov chain approximation and stochastic approximation-based iterative deep reinforcement learning algorithm where the probability distribution of the optimal stochastic controls is approximated by neural networks. In our hybrid algorithm, both Markov chain approximation and stochastic approximation are adopted in the learning processes. The idea of using the Markov chain approximation method to find initial guesses is proposed. A stochastic approximation is adopted to estimate the parameters of neural networks. Convergence analysis of the algorithm is presented. Numerical examples are provided to illustrate the performance of the algorithm.
Speaker
Zhuo Jin received his B.S. in Mathematics from Huazhong University of Science and Technology in 2005 and his Ph.D. in Mathematics from Wayne State University in 2011 (under George Yin). He is an Associate of the Society of Actuaries (ASA). Currently, he is a professor at the Department of Actuarial Studies and Business Analytics at Macquarie University in Australia. Before moving to Macquarie University in 2022, he was a faculty member at the Centre for Actuarial Studies, Department of Economics, The University of Melbourne for 10 years. His research interests include stochastic optimal control, numerical methods for stochastic systems, stochastic games, dividend optimization, retirement planning, portfolio selection, optimal reinsurance, model ambiguity, machine learning, reinforcement learning, systemic risk, cyber risk, pandemic risk, and climate risk. His publications appear in most of the major actuarial science journals and some prestigious journals on control, system science and OR.