Optimal Adaptive Allocation Using Deep Reinforcement Learning
Adjust Significance Level on a Simulation Basis
Allocation Rule Class
Clean the Python Virtual Environment
Build an Optimal Adaptive Allocation Rule using Reinforcement Learning
Configuration of Reinforcement Learning
DNN Configuration for Reinforcement Learning
Setting up a Python Virtual Environment
Simulate One Trial Using an Obtained Optimal Adaptive Allocation Rule
An implementation to compute an optimal adaptive allocation rule using deep reinforcement learning in a dose-response study (Matsuura et al. (2022) <doi:10.1002/sim.9247>). The adaptive allocation rule can directly optimize a performance metric, such as power, accuracy of the estimated target dose, or mean absolute error over the estimated dose-response curve.
Useful links