Sergey Leonov

Comparing Stochastic Training Environments for Deep Reinforcement Learning with Deep Q-network: Learning Stability and Transfer to Real Data

The paper investigates how the choice of a generative stochastic model of the training environment affects the numerical stability of the learning and the transfer of the Double DQN policy to real time series. Four environments — GBM, GARCH(1,1)-t, Minimal Neural SDE, and Neural Jump-Diffusion — are compared, created o...