Reproduce “Adaptive Weighted Nonnegative Matrix Factorization for Robust Feature Representation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Tingting Shen, Junhang Li, Can Tong, Qiang He, Chen Li, Yudong Yao, Yueyang Teng
The authors propose adaptive weighted NMF methods that learn a reliability weight for each sample during factorization, reducing the influence of poorly reconstructed samples. FWRNMF and EWRNMF alternate weight and factor updates. With Gaussian noise at c = 0.05, both reach 89.71% ACC on Page-blocks, versus 38.71% for NMFL2, but this is a k-means result under synthetic noise, not validated outlier detection.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.