Rethinking Non-Negative Matrix Factorization with Implicit Neural Representations

Krishna Subramani, Paris Smaragdis, Takuya Higuchi, Mehrez Souden
4/5/2024
eess.AScs.LGcs.SD

No official code found. Best implementation: SubramaniKrishna/in-nmf (5 stars, MIT, last commit 2024-03-24).

Abstract

Non-negative Matrix Factorization (NMF) is a powerful technique for analyzing regularly-sampled data, i.e., data that can be stored in a matrix. For audio, this has led to numerous applications using time-frequency (TF) representations like the Short-Time Fourier Transform. However extending these applications to irregularly-spaced TF representations, like the Constant-Q transform, wavelets, or sinusoidal analysis models, has not been possible since these representations cannot be directly stored in matrix form. In this paper, we formulate NMF in terms of learnable functions (instead of vectors) and show that NMF can be extended to a wider variety of signal classes that need not be regularly sampled.

View on arXivView PDF

Code Implementations(1)

SubramaniKrishna/in-nmfCommunity100%
MIT

Implicit Neural NMF

50Python, Jupyter NotebookFeb 23, 20242 years agoMIT

Similar Papers

Tackling Interpretability in Audio Classification Networks with Non-negative Matrix Factorization

80% match2023

Lee and Seung (2000)'s Algorithms for Non-negative Matrix Factorization: A Supplementary Proof Guide

80% match2025

Including local feature interactions in deep non-negative matrix factorization networks improves performance

79% match1 repo2025

Adversarial Generative NMF for Single Channel Source Separation

78% match2023

Applying non-negative matrix factorization with covariates to label matrix for classification

77% match2025

Non-Negative Matrix Factorization Using Non-Von Neumann Computers

77% match1 repo2025

Cite this paper

@article{subramani2024rethinking,
  title  = {Rethinking Non-Negative Matrix Factorization with Implicit Neural Representations},
  author = {Krishna Subramani and Paris Smaragdis and Takuya Higuchi and Mehrez Souden},
  year   = {2024},
  eprint = {2404.04439},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2404.04439v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2404.04439.svg)](https://codeofpaper.com/paper/2404.04439)

Discussion