Active Deep Probabilistic Subsampling

Hans van Gorp · Iris Huijben · Bastiaan Veeling · Nicola Pezzotti · Ruud J. G. van Sloun

Keywords: [ Sparsity and Compressed Sensing ] [ Algorithms ]

[ Abstract ]
[ Paper ]
[ Visit Poster at Spot B5 in Virtual World ]
Wed 21 Jul 9 a.m. PDT — 11 a.m. PDT
Spotlight presentation: Sparsity and Compressed Sensing
Wed 21 Jul 7 a.m. PDT — 8 a.m. PDT


Subsampling a signal of interest can reduce costly data transfer, battery drain, radiation exposure and acquisition time in a wide range of problems. The recently proposed Deep Probabilistic Subsampling (DPS) method effectively integrates subsampling in an end-to-end deep learning model, but learns a static pattern for all datapoints. We generalize DPS to a sequential method that actively picks the next sample based on the information acquired so far; dubbed Active-DPS (A-DPS). We validate that A-DPS improves over DPS for MNIST classification at high subsampling rates. Moreover, we demonstrate strong performance in active acquisition Magnetic Resonance Image (MRI) reconstruction, outperforming DPS and other deep learning methods.

Chat is not available.