Unsupervised Dimension Selection Using a Blue Noise Graph Spectrum

Published on May 1, 2019 in ICASSP (International Conference on Acoustics, Speech, and Signal Processing)
· DOI :10.1109/ICASSP.2019.8682551
Jayaraman J. Thiagarajan15
Estimated H-index: 15
(LLNL: Lawrence Livermore National Laboratory),
Rushil Anirudh8
Estimated H-index: 8
(LLNL: Lawrence Livermore National Laboratory)
+ 1 AuthorsPeer-Timo Bremer29
Estimated H-index: 29
(LLNL: Lawrence Livermore National Laboratory)
Unsupervised dimension selection is an important problem that seeks to reduce dimensionality of data, while preserving the most useful characteristics. While dimensionality reduction is commonly utilized to construct low-dimensional embeddings, they produce feature spaces that are hard to interpret. Further, in applications such as sensor design, one needs to perform reduction directly in the input domain, instead of constructing transformed spaces. Consequently, dimension selection (DS) aims to solve the combinatorial problem of identifying the top-k dimensions, which is required for effective experiment design, reducing data while keeping it interpretable, and designing better sensing mechanisms. In this paper, we develop a novel approach for DS based on graph signal analysis to measure feature influence. By analyzing synthetic graph signals with a blue noise spectrum, we show that we can measure the importance of each dimension. Using experiments in supervised learning and image masking, we demonstrate the superiority of the proposed approach over existing techniques in capturing crucial characteristics of high dimensional spaces, using only a small subset of the original features.
  • References (13)
  • Citations (0)
📖 Papers frequently viewed together
2019ICASSP: International Conference on Acoustics, Speech, and Signal Processing
5 Authors (Yong Peng, ..., Andrzej Cichocki)
1 Citations
6 Authors (Shichao Zhang, ..., Yonggang Li)
78% of Scinapse members use related papers. After signing in, all features are FREE.
#1Randal S. Olson (UPenn: University of Pennsylvania)H-Index: 16
#2William La Cava (UPenn: University of Pennsylvania)H-Index: 12
Last. Jason H. Moore (UPenn: University of Pennsylvania)H-Index: 72
view all 5 authors...
Background The selection, development, or comparison of machine learning methods in data mining can be a difficult task based on the target problem and goals of a particular study. Numerous publicly available real-world and simulated benchmark datasets have emerged from different sources, but their organization and adoption as standards have been inconsistent. As such, selecting and curating specific benchmarks remains an unnecessary burden on machine learning practitioners and data scientists.
40 CitationsSource
Nov 11, 2016 in SIGGRAPH (International Conference on Computer Graphics and Interactive Techniques)
#1Bhavya Kailkhura (LLNL: Lawrence Livermore National Laboratory)H-Index: 14
#2Jayaraman J. Thiagarajan (LLNL: Lawrence Livermore National Laboratory)H-Index: 15
Last. Pramod K. Varshney (SU: Syracuse University)H-Index: 66
view all 4 authors...
A common solution to reducing visible aliasing artifacts in image reconstruction is to employ sampling patterns with a blue noise power spectrum. These sampling patterns can prevent discernible artifacts by replacing them with incoherent noise. Here, we propose a new family of blue noise distributions, Stair blue noise, which is mathematically tractable and enables parameter optimization to obtain the optimal sampling distribution. Furthermore, for a given sample budget, the proposed blue noise ...
8 CitationsSource
#1Hamid Dadkhahi (UMass: University of Massachusetts Amherst)H-Index: 4
#2Marco F. Duarte (UMass: University of Massachusetts Amherst)H-Index: 36
We consider the problem of selecting an optimal mask for an image manifold, i.e., choosing a subset of the pixels of the image that preserves the manifold’s geometric structure present in the original data. Such masking implements a form of compressive sensing through emerging imaging sensor platforms for which the power expense grows with the number of pixels acquired. Our goal is for the manifold learned from masked images to resemble its full image counterpart as closely as possible. More pre...
4 CitationsSource
#1Siheng Chen (CMU: Carnegie Mellon University)H-Index: 17
#2Rohan Varma (CMU: Carnegie Mellon University)H-Index: 4
Last. Jelena Kovacevic (CMU: Carnegie Mellon University)H-Index: 39
view all 4 authors...
We propose a sampling theory for signals that are supported on either directed or undirected graphs. The theory follows the same paradigm as classical sampling theory. We show that perfect recovery is possible for graph signals bandlimited under the graph Fourier transform. The sampled signal coefficients form a new graph signal, whose corresponding graph structure preserves the first-order difference of the original graph signal. For general graphs, an optimal sampling operator based on experim...
263 CitationsSource
Apr 19, 2015 in ICASSP (International Conference on Acoustics, Speech, and Signal Processing)
#1Hamid Dadkhahi (UMass: University of Massachusetts Amherst)H-Index: 4
#2Marco F. Duarte (UMass: University of Massachusetts Amherst)H-Index: 36
We consider the problem of selecting a subset of the dimensions of an image manifold that best preserves the underlying local structure in the original data. We have previously shown that masks which preserve the data neighborhood graph are well suited to global manifold learning algorithms. However, local manifold learning algorithms leverage a geometric structure beyond that captured by this neighborhood graph. In this paper, we present a mask selection algorithm that further preserves this ad...
4 CitationsSource
#1Jayaraman J. Thiagarajan (LLNL: Lawrence Livermore National Laboratory)H-Index: 15
Last. Andreas SpaniasH-Index: 29
view all 4 authors...
Image understanding has been playing an increasingly crucial role in several inverse problems and computer vision. Sparse models form an important component in image understanding, since they emulate the activity of neural receptors in the primary visual cortex of the human brain. Sparse methods have been utilized in several learning problems because of their ability to provide parsimonious, interpretable, and efficient models. Exploiting the sparsity of natural signals has led to advances in se...
14 Citations
#1Daniel Heck (University of Konstanz)H-Index: 2
#2Thomas Schlömer (University of Konstanz)H-Index: 7
Last. Oliver Deussen (University of Konstanz)H-Index: 38
view all 3 authors...
In this article we revisit the problem of blue noise sampling with a strong focus on the spectral properties of the sampling patterns. Starting from the observation that oscillations in the power spectrum of a sampling pattern can cause aliasing artifacts in the resulting images, we synthesize two new types of blue noise patterns: step blue noise with a power spectrum in the form of a step function and single-peak blue noise with a wide zero-region and no oscillations except for a single peak. W...
40 CitationsSource
#1Aliaksei Sandryhaila (CMU: Carnegie Mellon University)H-Index: 19
#2Jose M. F. Moura (CMU: Carnegie Mellon University)H-Index: 61
In social settings, individuals interact through webs of relationships. Each individual is a node in a complex network (or graph) of interdependencies and generates data, lots of data. We label the data by its source, or formally stated, we index the data by the nodes of the graph. The resulting signals (data indexed by the nodes) are far removed from time or image signals indexed by well ordered time samples or pixels. DSP, discrete signal processing, provides a comprehensive, elegant, and effi...
612 CitationsSource
#1Zheng Zhao (SAS: SAS Institute)H-Index: 30
#2Lei Wang (UOW: University of Wollongong)H-Index: 38
Last. Jieping Ye (ASU: Arizona State University)H-Index: 62
view all 4 authors...
In the literature of feature selection, different criteria have been proposed to evaluate the goodness of features. In our investigation, we notice that a number of existing selection criteria implicitly select features that preserve sample similarity, and can be unified under a common framework. We further point out that any feature selection criteria covered by this framework cannot handle redundant features, a common drawback of these criteria. Motivated by these observations, we propose a ne...
191 CitationsSource
Jun 20, 2007 in ICML (International Conference on Machine Learning)
#1Zheng Zhao (ASU: Arizona State University)H-Index: 30
#2Huan Liu (ASU: Arizona State University)H-Index: 88
Feature selection aims to reduce dimensionality for building comprehensible learning models with good generalization performance. Feature selection algorithms are largely studied separately according to the type of learning: supervised or unsupervised. This work exploits intrinsic properties underlying supervised and unsupervised feature selection algorithms, and proposes a unified framework for feature selection based on spectral graph theory. The proposed framework is able to generate families...
526 CitationsSource
Cited By0