4418f6a54f4314202688d77956e731ce-Supplemental-Conference.pdf

Neural Information Processing Systems 

Forward propagation of the sparse coding layer is carried out by solving the optimization problem in(4). In this paper, we adopt the fast iterative shrinkage thresholding algorithm(FISTA)[1]. The results are provided in Figure B.1. In Figure C.1, we provide a visualization of the learned dictionary in the first layer of SDNet-18 trainedonImageNet.