Deep Learning for Continuous-time Stochastic Control with Jumps
This addresses stochastic control problems for researchers and practitioners, but it appears incremental as it adapts existing deep-learning methods to a specific class of problems.
The paper tackles finite-horizon continuous-time stochastic control problems with jumps by introducing a model-based deep-learning approach that trains neural networks for the optimal policy and value function, achieving accuracy and scalability in complex, high-dimensional tasks.
In this paper, we introduce a model-based deep-learning approach to solve finite-horizon continuous-time stochastic control problems with jumps. We iteratively train two neural networks: one to represent the optimal policy and the other to approximate the value function. Leveraging a continuous-time version of the dynamic programming principle, we derive two different training objectives based on the Hamilton-Jacobi-Bellman equation, ensuring that the networks capture the underlying stochastic dynamics. Empirical evaluations on different problems illustrate the accuracy and scalability of our approach, demonstrating its effectiveness in solving complex, high-dimensional stochastic control tasks.