Abstract | ||
---|---|---|
In this paper we propose a new LP-based formulation to solve near separable non negative matrix factorization (NMF) problem using L1 norm optimization. We also present a comprehensive experimental evaluation of the existing methods to solve separable NMF problem and compare them with the proposed formulation. The evaluation of this formulation on synthetic data shows that our new formulation gives significantly better quality of factorization as compared to the existing methods.
|
Year | DOI | Venue |
---|---|---|
2019 | 10.1145/3297001.3297016 | COMAD/CODS |
Keywords | Field | DocType |
NMF, Non-negative Matrix Factorisation | Applied mathematics,Separable space,Synthetic data,Factorization,Non-negative matrix factorization,Mathematics | Conference |
Citations | PageRank | References |
0 | 0.34 | 16 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Aashish Nagpal | 1 | 0 | 0.34 |
Chayan Sharma | 2 | 0 | 0.34 |
Rahul Garg | 3 | 884 | 85.42 |
Pawan Kumar | 4 | 1 | 1.64 |