MA7b.2

Reinforcement learning for sequential experimental design via actor-accelerated policy dual averaging

Caleb Ju, Georgia Institute of Technology, United States
Session:
MA7b: Modern approaches to experimental design Lecture
Track:
Adaptive Systems, Machine Learning, and Data Analytics
Location:
Chapel
Presentation Time:
Mon, 26 Oct, 10:40 - 11:05 PT (UTC -7)
Session Chair:
Wendy Di, Argonne National Laboratory, United States
Presentation
Discussion
Resources
No resources available.
Session MA7b
MA7b.1: Bayesian Optimal Experimental Design for Measurement Selection in Kalman Filtering
Xun Huan, University of Michigan, United States
MA7b.2: Reinforcement learning for sequential experimental design via actor-accelerated policy dual averaging
Caleb Ju, Georgia Institute of Technology, United States
MA7b.3: Exchange algorithm strategies for optimal experimental design
Srinivas Eswar, Vishwas Rao, Argonne National Laboratory, United States; Arvind Saibaba, North Carolina State University, United States
MA7b.4: Enabling Scalable Experimental Design for Dynamical Systems via Fast Bayesian Inverse Solvers
Yuanzhe Xi, Emory University, United States; Difeng Cai, Southern Methodist University, United States; Tianshi Xu, University of Kentucky, United States
Contacts