Markov Process

Markov processes model systems evolving through a sequence of states, where the future state depends only on the current state (the Markov property). Current research focuses on extending Markov process applications to diverse fields, including generative modeling (using piecewise deterministic Markov processes and autoregressive transformers), autonomous driving (scenario generation via Markov process editing), and biomolecular dynamics (using graph neural networks and variational approaches). These advancements improve the efficiency and accuracy of modeling complex systems, leading to better insights in various scientific domains and practical applications like anomaly detection and treatment effect estimation.