Continuous Time Stochastic

Continuous-time stochastic processes model systems evolving randomly over continuous time, aiming to understand and predict their behavior. Current research focuses on developing efficient inference methods for these processes, employing techniques like Monte Carlo Tree Search, neural networks (including graph neural networks and neural ordinary differential equations), and variational inference, often applied to problems involving Markov jump processes and stochastic differential equations. These advancements improve the accuracy and efficiency of modeling diverse phenomena, from molecular dynamics and financial markets to user behavior on digital platforms and control systems, impacting fields ranging from machine learning to systems biology.