Actuarial Science Seminar: A hybrid deep learning approach for optimal insurance strategies

This work studies a deep learning approach to find optimal insurance strategies for insurance companies. Due to the randomness of the financial ruin time to terminate the control processes, a Markov chain approximation-based iterative deep learning algorithm is developed to study this type of infinite-horizon optimal control problems. The optimal controls are approximated as deep neural networks. The framework of Markov chain approximation plays a key role in building the iterative equations and initialization of the algorithm. Optimal parameters of neural networks are then obtained iteratively. Convergence of the algorithm is studied. Satisfactory computation efficiency and accuracy are achieved as presented in numerical examples.
* zoom link available upon request