Machine Learning in Financial Mathematics with Michael Ludkovski

Explore optimal order execution in financial markets through a 57-minute virtual talk by Michael Ludkovski from the University of California, Santa Barbara. Delve into "Machine Learning Surrogates for Parametric and Adaptive Optimal Execution" as part of the SIAM Activity Group on Financial Mathematics and Engineering Virtual Talk Series. Learn about a novel model for discrete time, stochastic transient price impact, and discover how dynamic programming and deep learning are combined to create optimal strategies across various model configurations. Examine the innovative adaptive robust stochastic control framework that dynamically learns model parameters and incorporates Bayesian uncertainty. Witness the integration of dynamic learning, robustness, and control in this cutting-edge modeling approach. Gain insights into the 7-dimensional adaptive robust optimal order execution problem and compare it with alternative strategies. The talk concludes with a Q&A session, offering an opportunity to engage with the speaker and deepen your understanding of machine learning applications in financial mathematics.