1: \begin{abstract}

2: 			Nonnegative matrix factorization (NMF) has been widely studied in recent years due to its effectiveness in representing nonnegative data with parts-based representations. 

3: 			For NMF, a sparser solution implies better parts-based representation.

4: 			However, current NMF methods do not always generate sparse solutions.

5: 			In this paper, we propose a new NMF method with log-norm imposed on the factor matrices to enhance the sparseness.

6: 			Moreover, we propose a novel column-wisely sparse norm, named $\ell_{2,\log}$-(pseudo) norm to enhance the robustness of the proposed method.

7: 			The $\ell_{2,\log}$-(pseudo) norm is invariant, continuous, and differentiable.

8: 			For the $\ell_{2,\log}$ regularized shrinkage problem, we derive a closed-form solution, which can be used for other general problems.

9: 			Efficient multiplicative updating rules are developed for the optimization, which theoretically guarantees the convergence of the objective value sequence.

10: 			Extensive experimental results confirm the effectiveness of the proposed method, as well as the enhanced sparseness and robustness.

11: 		\end{abstract}

12: