Researchers have developed a novel Newton-type algorithm for Nonnegative Matrix Factorization (NMF) that utilizes the Kullback-Leibler (KL) divergence. This new method offers an efficient approach for analyzing count datasets, such as term-document matrices and images, by employing a second-order Taylor expansion of the loss function. The algorithm, which generalizes the HALS algorithm, has demonstrated provable convergence and competitive performance against existing state-of-the-art methods across various datasets. AI
IMPACT This research introduces a more efficient algorithm for NMF, potentially improving performance on count-based data analysis tasks in machine learning.
RANK_REASON The cluster contains a research paper detailing a new algorithm for a machine learning task.
- alphaXiv
- arXiv
- CatalyzeX
- DagsHub
- Gotit.pub
- HALS algorithm
- Hugging Face
- IArxiv
- image
- Kullback--Leibler divergence
- Newton's method
- non-negative matrix factorization
- Poisson distribution
- ScienceCast
- term-document matrices
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
