arXiv subjects

Keunsu Kim

Publications and source records attributed to Keunsu Kim.

5 recordsLinked to original sources

arXiv2026

Non-negative Matrix Factorisation with Topological Regularisation

Non-negative matrix factorisation (NMF) learns additive representations from data, but non-negativity alone does not ensure interpretable basis functions. We introduce Top-NMF, which guides basis learning through general topological preferences without prescribing the detailed form of the components. Observations and basis vectors are treated as non-negative functions on structured domains. Discrete topological conditions are difficult to optimise directly. Our guiding viewpoint is that persistent homology provides a stable, continuous relaxation of ordinary homological invariants: it tracks connected components and loops across thresholds, allowing structural preferences to be expressed through continuous scores suitable for optimisation. We construct such scores for connected image parts, clique-like graph structure, and periodic time-series components, and incorporate them into a common NMF objective. For graph data, we derive an exact maximum-weight-spanning-tree formula and characterise all local minimisers of the score, showing that their positive-weight edges form disjoint cliques. We establish the regularity of the scores, describe their derivatives, and analyse projected first-order optimisation. Numerical studies on synthetic and real data demonstrate how these priors guide basis learning and examine the balance between structural agreement, atom recovery, and reconstruction accuracy.

cs.LG↗

arXiv2025

Exact multi-parameter persistent homology of time-series data: Fast and variable topological inferences

We propose a novel exact multi-parameter persistent homology method for analyzing time-series data utilizing the Liouville torus. In the field of topological data analysis (TDA), the conventional approach to analyzing time-series data often involves sliding window embedding. From the perspective of Takens' embedding theorem, we justify the analysis of the Liouville torus in TDA and discuss the similarities and differences between the Liouville torus and sliding window embedding approaches. We develop a multi-parameter filtration method based on Fourier decomposition and provide an exact formula of persistent homology with its one-parameter reduction of the multi-parameter filtration. The conventional TDA of time-series data via sliding window is known to be computationally expensive, but the proposed method yields the exact barcode formula with the symmetry of the Liouville torus promptly, which significantly reduces computational complexity while demonstrating comparable or superior performance compared to the existing TDA methods. Furthermore, the proposed method provides a way of obtaining various topological inferences by exploring different filtration rays within the multi-parameter filtration space, utilizing the nearly real-time computational capabilities of the proposed method. The advantages of the proposed method significantly improve the efficiency and flexibility of TDA when handling extensive time-series data within machine learning workflows.

math.AT↗

arXiv2024

Filtration learning in exact multi-parameter persistent homology and classification of time-series data

To analyze the topological properties of the given discrete data, one needs to consider a continuous transform called filtration. Persistent homology serves as a tool to track changes of homology in the filtration. The outcome of the topological analysis of data varies depending on the choice of filtration, making the selection of filtration crucial. Filtration learning is an attempt to find an optimal filtration that minimizes the loss function. Exact Multi-parameter Persistent Homology (EMPH) has been recently proposed, particularly for topological time-series analysis, that utilizes the exact formula of rank invariant instead of calculating it. In this paper, we propose a framework for filtration learning of EMPH. We formulate an optimization problem and propose an algorithm for solving the problem. We then apply the proposed algorithm to several classification problems. Particularly, we derive the exact formula of the gradient of the loss function with respect to the filtration parameters, which makes it possible to directly update the filtration without using automatic differentiation, significantly enhancing the learning process.

math.OC↗

arXiv2024

Supervised low-rank semi-nonnegative matrix factorization with frequency regularization for forecasting spatio-temporal data

We propose a novel methodology for forecasting spatio-temporal data using supervised semi-nonnegative matrix factorization (SSNMF) with frequency regularization. Matrix factorization is employed to decompose spatio-temporal data into spatial and temporal components. To improve clarity in the temporal patterns, we introduce a nonnegativity constraint on the time domain along with regularization in the frequency domain. Specifically, regularization in the frequency domain involves selecting features in the frequency space, making an interpretation in the frequency domain more convenient. We propose two methods in the frequency domain: soft and hard regularizations, and provide convergence guarantees to first-order stationary points of the corresponding constrained optimization problem. While our primary motivation stems from geophysical data analysis based on GRACE (Gravity Recovery and Climate Experiment) data, our methodology has the potential for wider application. Consequently, when applying our methodology to GRACE data, we find that the results with the proposed methodology are comparable to previous research in the field of geophysical sciences but offer clearer interpretability.

stat.ML↗

arXiv2020

Highly-scalable stochastic neuron based on Ovonic Threshold Switch (OTS) and its applications in Restricted Boltzmann Machine (RBM)

Interest in Restricted Boltzmann Machine (RBM) is growing as a generative stochastic artificial neural network to implement a novel energy-efficient machine-learning (ML) technique. For a hardware implementation of the RBM, an essential building block is a reliable stochastic binary neuron device that generates random spikes following the Boltzmann distribution. Here, we propose a highly-scalable stochastic neuron device based on Ovonic Threshold Switch (OTS) which utilizes the random emission and capture process of traps as the source of stochasticity. The switching probability is well described by the Boltzmann distribution, which can be controlled by operating parameters. As a candidate for a true random number generator (TRNG), it passes 15 among the 16 tests of the National Institute of Standards and Technology (NIST) Statistical Test Suite (Special Publication 800-22). In addition, the recognition task of handwritten digits (MNIST) is demonstrated using a simulated RBM network consisting of the proposed device with a maximum recognition accuracy of 86.07 %. Furthermore, reconstruction of images is successfully demonstrated using images contaminated with noises, resulting in images with the noise removed. These results show the promising properties of OTS-based stochastic neuron devices for applications in RBM systems.

physics.app-ph↗