Home
Portal
Grstats
Latest images
Register
Log in
grstats- Posts : 959
Join date : 2009-10-21 
AUEB Stats Seminars 4/3/2021: Practical Distributionally Robust Markov Decision Processes using Relative Entropy by William Greenall (UCL)
Tue 2 Mar 2021 - 22:15
AUEB STATISTICS SEMINAR SERIES MARCH 2021
William Greenall (PhD student, UCL. Supervisor: Petros Dellaportas)
Practical Distributionally Robust Markov Decision Processes using Relative Entropy
ABSTRACT
Distributionally Robust Markov Decision Processes offer a toolset for improving performance of sequential optimisation algorithms in the face of poor or particularly uncertain estimates of a transition model. The literature has focused on the use of Wasserstein distances as a tool for regulating the extent of robustness, but is not simple to use due to its lack of closed forms. On the other hand, the Kullback-Leibler divergence has been shunned as its use has heretofore implied limited flexibility. I present a method to render the KL-divergence a useful and practical tool for the construction of ambiguity sets, and build both discrete-state and continuous-state space decision processes using the formulation.
link: [url=meet.google.com/ywg-hzrp-xgf]meet.google.com/ywg-hzrp-xgf[/url]
Ημερομηνία Εκδήλωσης:
Thursday, March 4, 2021 - 12:30
William Greenall (PhD student, UCL. Supervisor: Petros Dellaportas)
Practical Distributionally Robust Markov Decision Processes using Relative Entropy
ABSTRACT
Distributionally Robust Markov Decision Processes offer a toolset for improving performance of sequential optimisation algorithms in the face of poor or particularly uncertain estimates of a transition model. The literature has focused on the use of Wasserstein distances as a tool for regulating the extent of robustness, but is not simple to use due to its lack of closed forms. On the other hand, the Kullback-Leibler divergence has been shunned as its use has heretofore implied limited flexibility. I present a method to render the KL-divergence a useful and practical tool for the construction of ambiguity sets, and build both discrete-state and continuous-state space decision processes using the formulation.
link: [url=meet.google.com/ywg-hzrp-xgf]meet.google.com/ywg-hzrp-xgf[/url]
Ημερομηνία Εκδήλωσης:
Thursday, March 4, 2021 - 12:30
- AUEB Stats Seminars 26/11/2021: A new framework of semi-Markov processes for parameter estimation and Reliability Analysis by Andreas Makrides (University of the Aegean)
- AUEB Stats Seminars 2/12/2022: Federated Bayesian Computation via Piecewise Deterministic Markov Processes (Delft Institute of Applied Mathematics, NL)
- AUEB Stats Seminars 5/2/2021: Manifold Markov chain Monte Carlo methods for Bayesian inference in a wide class of diffusion models
- AUEB Stats Seminars 19/11/2021: Estimation of optimal individualized treatment rules for multistate disease processes by Giorgos Bakoyannis (Indiana University School of Medicine)
- AUEB Stats Seminars 27/5/2021: Manifold Markov chain Monte Carlo methods for Bayesian inference in a wide class of diffusion models by Alexandros Beskos
Permissions in this forum:
You cannot reply to topics in this forum|
|
|