Fast greedy algorithms for dictionary selection with generalized sparsity constraints
This work addresses dictionary selection for data approximation, offering a faster and more flexible solution, though it appears incremental as it builds on greedy algorithms with new constraints.
The authors tackled the problem of dictionary selection by proposing a new greedy algorithm that is faster than existing methods and handles complex sparsity constraints like average sparsity. They demonstrated through experiments that it outperforms known methods and achieves competitive performance with dictionary learning algorithms in reduced running time.
In dictionary selection, several atoms are selected from finite candidates that successfully approximate given data points in the sparse representation. We propose a novel efficient greedy algorithm for dictionary selection. Not only does our algorithm work much faster than the known methods, but it can also handle more complex sparsity constraints, such as average sparsity. Using numerical experiments, we show that our algorithm outperforms the known methods for dictionary selection, achieving competitive performances with dictionary learning algorithms in a smaller running time.