Decision Making in Automated Experiments - From Bandits and Bayesian Optimization to Reinforcement Learning and Stochastic Optimization

Explore decision-making strategies for automated experiments through this tutorial covering multi-armed bandits, Bayesian optimization, reinforcement learning, and stochastic optimization techniques. Learn from Sergei V. Kalinin of the University of Tennessee, Knoxville, as he demonstrates how these computational approaches can enhance experimental design and data collection in physics research. Discover practical applications of machine learning algorithms for optimizing experimental parameters, making real-time decisions during data acquisition, and improving the efficiency of scientific investigations. Gain insights into how automated decision-making frameworks can accelerate discovery processes and reduce the time required for experimental optimization across various physics domains.