Timezone: »

Active Deep Probabilistic Subsampling
Hans van Gorp · Iris Huijben · Bastiaan Veeling · Nicola Pezzotti · Ruud J. G. van Sloun

Wed Jul 21 09:00 AM -- 11:00 AM (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.

Author Information

Hans van Gorp (Eindhoven University of Technology)
Iris Huijben (Eindhoven University of Technology)
Iris Huijben

I work on representation learning with a focus on discretization. We apply discretization either at the raw input signal in order to achieve hard attention or for compression purposes, or in the latent representation in order to facilitate pattern recognition.

Bastiaan Veeling (University of Amsterdam)
Nicola Pezzotti (Philips)
Ruud J. G. van Sloun (Technical university of Eindhoven)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors