We propose to learn an active learning policy for inverse problem optimisation.
We propose to learn a switching policy for inverse problem optimisation.
We use policy gradient methods to learn active acquisition policies for improving MRI reconstruction.