DRGraph: An Efficient Graph Layout Algorithm for Large-scale Graphs by Dimensionality Reduction

Dimensionality reduction refers to a set of mathematical techniques used to reduce complexity of the original high-dimensional data, while preserving its selected properties. Improvements in simulation strategies and experimental data collection methods are resulting in a deluge of heterogeneous and high-dimensional data, which often makes dimensionality reduction the only viable way to gain qualitative and quantitative understanding of the data. However, existing dimensionality reduction software often does not scale to datasets arising in real-life applications, which may consist of thousands of points with millions of dimensions. In this paper, we propose a parallel framework for dimensionality reduction of large-scale data. We identify key components underlying the spectral dimensionality reduction techniques, and propose their efficient parallel implementation. We show that the resulting framework can be used to process datasets consisting of millions of points when executed on a 16,000-core cluster, which is beyond the reach of currently available methods. To further demonstrate applicability of our framework we perform dimensionality reduction of 75,000 images representing morphology evolution during manufacturing of organic solar cells in order to identify how processing parameters affect morphology evolution.

Download Full-text

Dimensionality reduction using compressed sensing and its application to a large-scale visual recognition task

The 2010 International Joint Conference on Neural Networks (IJCNN) ◽

10.1109/ijcnn.2010.5596477 ◽

2010 ◽

Cited By ~ 4

Author(s):

Jie Yang ◽

Abdesselam Bouzerdoum ◽

Fok Hing Chi Tivive ◽

Son Lam Phung

Keyword(s):

Compressed Sensing ◽

Dimensionality Reduction ◽

Large Scale ◽

Visual Recognition ◽

Recognition Task

Download Full-text

A network information theoretic framework to characterise muscle synergies in space and time

10.1101/2021.10.15.464450 ◽

2021 ◽

Author(s):

David O'Reilly ◽

Ioannis Delis

Keyword(s):

Dimensionality Reduction ◽

Motor System ◽

Large Scale ◽

Muscle Synergy ◽

Whole Body ◽

Muscle Synergies ◽

Spatial And Temporal Scales ◽

Network Information ◽

Benchmark Datasets ◽

Muscle Activations

Coordinated movement is thought to be simplified by the nervous system through the activation of muscle synergies. Current approaches to muscle synergy extraction rely on dimensionality reduction algorithms that impose limiting constraints. To capture large-scale interactions between muscle activations, a more generalised approach that considers the complexity and nonlinearity of the motor system is required. Here we developed a novel framework for muscle synergy extraction that relaxes model assumptions by using a combination of information- and network theory and dimensionality reduction. This novel framework can capture spatial, temporal and spatiotemporal interactions, producing distinct spatial groupings and both tonic and phasic temporal patterns. Furthermore, our framework identifies submodular structures in the extracted synergies that exemplify the fractal modularity of the human motor system. To demonstrate the versatility of the methodology, we applied it to two benchmark datasets of arm and whole-body reaching movements. Readily interpretable muscle synergies spanning multiple spatial and temporal scales were identified that demonstrated significant task dependence, ability to capture trial-to-trial fluctuations, a scale-invariance with dataset complexity and a substantial concordance across participants. Finally, we position this framework as a bridge between existing models and recent network-theoretic endeavours by discussing the continuity and novelty of the presented findings.

Download Full-text

An Evaluation of Supervised Dimensionality Reduction For Large Scale Data

Journal of Machine and Computing ◽

10.53759/7669/jmc202202003 ◽

2022 ◽

pp. 17-25

Author(s):

Nancy Jan Sliper

Keyword(s):

Dimensionality Reduction ◽

Large Scale ◽

Simulated Data ◽

Principal Component ◽

Low Rank ◽

Learning Tools ◽

Large Scale Data ◽

Reduction Methods ◽

Low Dimensional ◽

Scale Data

Experimenters today frequently quantify millions or even billions of characteristics (measurements) each sample to address critical biological issues, in the hopes that machine learning tools would be able to make correct data-driven judgments. An efficient analysis requires a low-dimensional representation that preserves the differentiating features in data whose size and complexity are orders of magnitude apart (e.g., if a certain ailment is present in the person's body). While there are several systems that can handle millions of variables and yet have strong empirical and conceptual guarantees, there are few that can be clearly understood. This research presents an evaluation of supervised dimensionality reduction for large scale data. We provide a methodology for expanding Principal Component Analysis (PCA) by including category moment estimations in low-dimensional projections. Linear Optimum Low-Rank (LOLR) projection, the cheapest variant, includes the class-conditional means. We show that LOLR projections and its extensions enhance representations of data for future classifications while retaining computing flexibility and reliability using both experimental and simulated data benchmark. When it comes to accuracy, LOLR prediction outperforms other modular linear dimension reduction methods that require much longer computation times on conventional computers. LOLR uses more than 150 million attributes in brain image processing datasets, and many genome sequencing datasets have more than half a million attributes.

Download Full-text