Showing 1–3 of 3 results for author: Kawasumi, R

Automatic Hyperparameter Tuning in Sparse Matrix Factorization

Authors: Ryota Kawasumi, Koujin Takeda

Abstract: We study the problem of hyperparameter tuning in sparse matrix factorization under Bayesian framework. In the prior work, an analytical solution of sparse matrix factorization with Laplace prior was obtained by variational Bayes method under several approximations. Based on this solution, we propose a novel numerical method of hyperparameter tuning by evaluating the zero point of normalization fac… ▽ More We study the problem of hyperparameter tuning in sparse matrix factorization under Bayesian framework. In the prior work, an analytical solution of sparse matrix factorization with Laplace prior was obtained by variational Bayes method under several approximations. Based on this solution, we propose a novel numerical method of hyperparameter tuning by evaluating the zero point of normalization factor in sparse matrix prior. We also verify that our method shows excellent performance for ground-truth sparse matrix reconstruction by comparing it with the widely-used algorithm of sparse principal component analysis. △ Less

Submitted 17 May, 2023; originally announced May 2023.

Comments: 13 pages, 5 figures

Journal ref: Neural Computation (2023) 35 (6) 1086-1099

Pointwise convergence of Fourier series and deep neural network for the indicator function of d-dimensional ball

Authors: Ryota Kawasumi, Tsuyoshi Yoneda

Abstract: In this paper, we clarify the crucial difference between a deep neural network and the Fourier series. For the multiple Fourier series of periodization of some radial functions on $\mathbb{R}^d$, Kuratsubo (2010) investigated the behavior of the spherical partial sum and discovered the third phenomenon other than the well-known Gibbs-Wilbraham and Pinsky phenomena. In particular, the third one exh… ▽ More In this paper, we clarify the crucial difference between a deep neural network and the Fourier series. For the multiple Fourier series of periodization of some radial functions on $\mathbb{R}^d$, Kuratsubo (2010) investigated the behavior of the spherical partial sum and discovered the third phenomenon other than the well-known Gibbs-Wilbraham and Pinsky phenomena. In particular, the third one exhibits prevention of pointwise convergence. In contrast to it, we give a specific deep neural network and prove pointwise convergence. △ Less

Submitted 25 June, 2024; v1 submitted 17 April, 2023; originally announced April 2023.

Comments: When the version 2 was rejected (where I submitted it to an AI journal), I realized I needed to further clarify the key point, and also realized the field is rather Fourier analysis

Approximate Method of Variational Bayesian Matrix Factorization/Completion with Sparse Prior

Authors: Ryota Kawasumi, Koujin Takeda

Abstract: We derive analytical expression of matrix factorization/completion solution by variational Bayes method, under the assumption that observed matrix is originally the product of low-rank dense and sparse matrices with additive noise. We assume the prior of sparse matrix is Laplace distribution by taking matrix sparsity into consideration. Then we use several approximations for derivation of matrix f… ▽ More We derive analytical expression of matrix factorization/completion solution by variational Bayes method, under the assumption that observed matrix is originally the product of low-rank dense and sparse matrices with additive noise. We assume the prior of sparse matrix is Laplace distribution by taking matrix sparsity into consideration. Then we use several approximations for derivation of matrix factorization/completion solution. By our solution, we also numerically evaluate the performance of sparse matrix reconstruction in matrix factorization, and completion of missing matrix element in matrix completion. △ Less

Submitted 14 March, 2018; originally announced March 2018.

Comments: 22 pages, 4 figures, part of this work was presented in IEEE International Workshop on Machine Learning for Signal Processing (2017)

Journal ref: J. Stat. Mech. (2018) 053404