Non-negative matrix factorization (NMF) known as learnt parts-based representation has become a data analysis tool for clustering tasks. It provides an alternative learning paradigm to cope with non-negative data clus...
详细信息
Non-negative matrix factorization (NMF) known as learnt parts-based representation has become a data analysis tool for clustering tasks. It provides an alternative learning paradigm to cope with non-negative data clustering. In this paradigm, concept factorization (CF) and symmetric non-negative factorization (SymNMF) are two typically important representative models. In general, they have distinct behaviors: in CF, each cluster is modeled as a linear combination of samples, and vice versa, i.e., sample reconstruction, while SymNMF built on pair-wise sample similarity measure, is to preserve similarity of samples in a low-dimensional subspace, namely similarity reconstruction. In this paper, we propose a similarity-based concept factorization (SCF) as a synthesis of the two behaviors. This design can be formulated as: the similarity of reconstructed samples by CF is close to that of original samples. To optimize it, we develop an optimization algorithm which leverages the alternating direction of multipliers (ADMM) method to solve each sub-problem of SCF. Besides, we take a further step to consider the robust issue of similarity reconstruction and explore a robust SCF model (RSCF), which penalizes the hardest pair-wise similarity reconstruction via l(infinity). Thus, RSCF enjoys similarity preservation, robustness to similarity perturbation, and ability of reconstructing samples. Extensive experiments validate such properties and show that the proposed SCF and RSCF achieve large performance gains as compared to their counterparts.
A hybrid system of cellular mode and device-to-device （D2D） mode is considered in this paper, where the cellular resource is reused by the D2D transmission. With the objective of capacity maximization, the power opt...
详细信息
A hybrid system of cellular mode and device-to-device （D2D） mode is considered in this paper, where the cellular resource is reused by the D2D transmission. With the objective of capacity maximization, the power optimization of D2D sub-system is considered, taking into account quality of service （QoS） requirement. The power optimization problem is divided into two stages： The first stage is the admission control scheme design based on the QoS requirement of D2D users, and the second is power allocation to maximize aggregate throughput of admissible D2D users. For the D2D admission control problem, a heuristic sorting-based algorithm is proposed to index the admissible D2D links, where gain to Interference ratio （GIR） sorting criterion is used. Applying an approximate form of Shannon capacity, the power allocation problem can be solved by convex optimization and geometric programming tools efficiently. based on the theoretical analysis, a practical algorithm is proposed. The precision can reach a trade-off between complexity and performance. Numerical simulation results confirm that combining with GIR sorting method, the proposed scheme can significantly improve the D2D system＇s capacity and fairness.
暂无评论