Split Slice Training Augmentation and Hyperparameter Tuning of RAKI Networks for Simultaneous Multi-Slice Reconstruction

2020/06/02 by Andrew S. Nencka, Nencka, Andrew S., PhD +18
Computer Science · Engineering · Physics and Astronomy · #Gait Recognition and Analysis #Hand Gesture Recognition Systems #Ultrasonics and Acoustic Wave Propagation #eess.IV #physics.med-ph

paper · pdf · doi:10.48550/arxiv.2006.01917

arxiv created 2020/06/02 · arxiv updated 2020/06/04

Abstract

Split-slice augmentation for simultaneous multi-slice RAKI networks positively impacts network performance. Hyperparameter tuning of such reconstruction networks can lead to further improvements in unaliasing performance.

Citations

• ImageNet classification with deep convolutional neural networks

Related

• Network Recasting: A Universal Method for Network Architecture Transformation
• Split-Boost Neural Networks
• Understanding and Optimizing Packed Neural Network Training for Hyper-Parameter Tuning
• Progressive Skeletonization: Trimming more fat from a network at initialization
• Practical Network Acceleration with Tiny Sets: Hypothesis, Theory, and Algorithm