Test-Time Training Can Close the Natural Distribution Shift Performance Gap in Deep Learning Based Compressed Sensing.
Mohammad Zalbagi DarestaniJiayu LiuReinhard HeckelPublished in: ICML (2022)
Keyphrases
- deep learning
- compressed sensing
- deep architectures
- image reconstruction
- random projections
- unsupervised learning
- sparse representation
- natural images
- training set
- signal processing
- supervised learning
- machine learning
- weakly supervised
- data sets
- mental models
- similarity measure
- computer vision
- dimensionality reduction
- text classification
- co occurrence
- training examples
- pattern recognition