IST A-NAS: Efficient and Consistent Neural Architecture Search by Sparse Coding Yibo Y ang

Current gradient-based NAS methods commonly ignore the constraint of sparsity in the search phase, but project the optimized solution onto a sparse one by post-processing.