
Deep Learning through Sparse Representation and Low-Rank Modeling bridges classical sparse and low rank models--those that emphasize problem-specific Interpretability--with recent deep network models that have enabled a larger learning capacity and better utilization...