no code implementations • 15 Feb 2023 • Matthew Werenski, Shoaib Bin Masud, James M. Murphy, Shuchin Aeron
This paper considers the use of recently proposed optimal transport-based multivariate test statistics, namely rank energy and its variant the soft rank energy derived from entropically regularized optimal transport, for the unsupervised nonparametric change point detection (CPD) problem.
no code implementations • 21 Oct 2022 • Marshall Mueller, Shuchin Aeron, James M. Murphy, Abiy Tasissa
We show this approach leads to sparse representations in Wasserstein space and addresses the problem of non-uniqueness of barycentric representation.
1 code implementation • 28 Apr 2022 • Kangning Cui, Ruoning Li, Sam L. Polk, James M. Murphy, Robert J. Plemmons, Raymond H. Chan
DSIRC then locates high-density, high-purity pixels far in diffusion distance (a data-dependent distance metric) from other high-density, high-purity pixels and treats these as cluster exemplars, giving each a unique label.
no code implementations • 19 Apr 2022 • Sam L. Polk, Aland H. Y. Chan, Kangning Cui, Robert J. Plemmons, David A. Coomes, James M. Murphy
Ash dieback (Hymenoscyphus fraxineus) is an introduced fungal disease that is causing the widespread death of ash trees across Europe.
1 code implementation • 13 Apr 2022 • Sam L. Polk, Kangning Cui, Robert J. Plemmons, James M. Murphy
Hyperspectral images encode rich structure that can be exploited for material discrimination by machine learning algorithms.
1 code implementation • 18 Mar 2022 • Sam L. Polk, Kangning Cui, Aland H. Y. Chan, David A. Coomes, Robert J. Plemmons, James M. Murphy
Hyperspectral images taken from aircraft or satellites contain information from hundreds of spectral bands, within which lie latent lower-dimensional structures that can be exploited for classifying vegetation and other materials.
1 code implementation • 28 Jan 2022 • Matthew Werenski, Ruijie Jiang, Abiy Tasissa, Shuchin Aeron, James M. Murphy
Our first main result leverages the Riemannian geometry of Wasserstein-2 space to provide a procedure for recovering the barycentric coordinates as the solution to a quadratic optimization problem assuming access to the true reference measures.
no code implementations • 29 Oct 2021 • Shoaib Bin Masud, Matthew Werenski, James M. Murphy, Shuchin Aeron
We leverage this result to demonstrate fast convergence of sample sRE and sRMMD to their population version making them useful for high-dimensional GoF testing.
1 code implementation • 29 Mar 2021 • Sam L. Polk, James M. Murphy
Clustering algorithms partition a dataset into groups of similar points.
1 code implementation • 31 Jan 2021 • James M. Murphy, Sam L. Polk
To efficiently learn the multiscale structure observed in many real datasets, we introduce the Multiscale Learning by Unsupervised Nonlinear Diffusion (M-LUND) clustering algorithm, which is derived from a diffusion process at a range of temporal scales.
1 code implementation • 3 Dec 2020 • Pranay Tankala, Abiy Tasissa, James M. Murphy, Demba Ba
We theoretically analyze the proposed program by relating the weighted $\ell_1$ penalty in KDS to a weighted $\ell_0$ program.
1 code implementation • 10 Apr 2020 • Shukun Zhang, James M. Murphy
We propose a method for the unsupervised clustering of hyperspectral images based on spatially regularized spectral clustering with ultrametric path distances.
no code implementations • 7 Mar 2020 • Lenore Cowen, Kapil Devkota, Xiaozhe Hu, James M. Murphy, Kaiyi Wu
This article develops a theory for DSD based on the multitemporal emergence of mesoscopic equilibria in the underlying diffusion process.
no code implementations • 6 Nov 2019 • James M. Murphy
An active learning algorithm for the classification of high-dimensional images is proposed in which spatially-regularized nonlinear diffusion geometry is used to characterize cluster cores.
no code implementations • 30 May 2019 • Mauro Maggioni, James M. Murphy
This article proposes an active learning method for high dimensional data, based on intrinsic data geometries learned through diffusion processes on graphs.
no code implementations • 8 Feb 2019 • James M. Murphy, Mauro Maggioni
The explicit incorporation of spatial regularity into the diffusion construction leads to smoother random processes that are more adapted for unsupervised machine learning than those based on spectra alone.
no code implementations • 15 Oct 2018 • Mauro Maggioni, James M. Murphy
This paper proposes and analyzes a novel clustering algorithm that combines graph-based diffusion geometry with techniques based on density and mode estimation.
1 code implementation • 17 Dec 2017 • Anna Little, Mauro Maggioni, James M. Murphy
We consider the problem of clustering with the longest-leg path distance (LLPD) metric, which is informative for elongated and irregularly shaped clusters.
no code implementations • 26 Apr 2017 • James M. Murphy, Mauro Maggioni
The problem of unsupervised learning and segmentation of hyperspectral images is a significant challenge in remote sensing.
no code implementations • 27 Feb 2016 • Wojciech Czaja, James M. Murphy, Daniel Weinberg
We justify the use of shearlets mathematically, before presenting a denoising single-image superresolution algorithm that combines the shearlet transform with sparse mixing estimators (SME).