Reinforcement learning

Experimental design for MRI by greedy policy search

We use policy gradient methods to learn active acquisition policies for improving MRI reconstruction.