paper-with-me

Papers

Leveraging Joint-Diagonalization in Transform-Learning NMF

2021-12-10 · Sixin Zhang, Emmanuel Soubies, Cédric Févotte

Non-negative matrix factorization with transform learning (TL-NMF) is a recent idea that aims at learning data representations suited to NMF. In this work, we relate TL-NMF to the classical matrix joint-diagonalization (JD) problem. We show that, when the number of data realizations is sufficiently large, TL-NMF can be replaced by a two-step approach -- termed as JD+NMF -- that estimates the transform through JD, prior to NMF computation. In contrast, we found that when the number of data realizations is limited, not only is JD+NMF no longer equivalent to TL-NMF, but the inherent low-rank constraint of TL-NMF turns out to be an essential ingredient to learn meaningful transforms for NMF.

📄 PDF Abstract BibTeX arXiv:2112.05664

Code (1)

sixin-zh/tlnmf-tsp 공식 구현

Similar Papers 제목 키워드 기반

Beyond Pham's algorithm for joint diagonalization

2018-11-28 · Pierre Ablin, Jean-François Cardoso, Alexandre Gramfort

The approximate joint diagonalization of a set of matrices consists in finding a basis in which these matrices are as diagonal as possible. This problem naturally appears in several statistical learning tasks such as bli…

Robust Blind Source Separation by Soft Decision-Directed Non-Unitary Joint Diagonalization

2021-06-28 · Wenjuan Liu, DaZheng Feng, Bingnan Pei, Mengdao Xing 외

Approximate joint diagonalization of a set of matrices provides a powerful framework for numerous statistical signal processing applications. For non-unitary joint diagonalization (NUJD) based on the least-squares (LS) c…

blind source separation

Identification of Matrix Joint Block Diagonalization

2020-11-02 · Yunfeng Cai, Ping Li

Given a set $\mathcal{C}=\{C_i\}_{i=1}^m$ of square matrices, the matrix blind joint block diagonalization problem (BJBDP) is to find a full column rank matrix $A$ such that $C_i=A\Sigma_iA^\text{T}$ for all $i$, where $…

Rethinking LDA: moment matching for discrete ICA

2015-07-07 · NeurIPS 2015 12 · Anastasia Podosinnikova, Francis Bach, Simon Lacoste-Julien

We consider moment matching techniques for estimation in Latent Dirichlet Allocation (LDA). By drawing explicit links between LDA and discrete versions of independent component analysis (ICA), we first derive a new set o…

Accelerating Matrix Diagonalization through Decision Transformers with Epsilon-Greedy Optimization

2024-06-23 · Kshitij Bhatta, Geigh Zollicoffer, Manish Bhattarai, Phil Romero 외

This paper introduces a novel framework for matrix diagonalization, recasting it as a sequential decision-making problem and applying the power of Decision Transformers (DTs). Our approach determines optimal pivot select…

Decision MakingSequential Decision MakingTransfer Learning