scholarly journals Streaming Algorithms for Subspace Analysis: an Edge- and Hardware-oriented review

2021 ◽  
Author(s):  
Alex Marchioni ◽  
Luciano Prono ◽  
Mauro Mangia ◽  
Fabio Pareschi ◽  
Riccardo Rovatti ◽  
...  
Keyword(s):  
High Dimensional ◽  
Streaming Algorithms ◽  
Subspace Analysis ◽  
Basic Tool ◽  
Hardware Platform ◽  
Common Edge ◽  
Redundant Data ◽  
Real World Applications ◽  
Early Processing ◽  
Digital Hardware

Subspace analysis is a basic tool for coping with high-dimensional data and is becoming a fundamental step in early processing of many signals elaboration tasks. Nowadays trend of collecting huge quantities of usually very redundant data by means of decentralized systems suggests these techniques be deployed as close as possible to the data sources. Regrettably, despite its conceptual simplicity, subspace analysis is ultimately equivalent to eigenspace computation and brings along non-negligible computational and memory requirements. To make this fit into typical systems operating at the edge, specialized streaming algorithms have been recently devised that we here classify and review giving them a coherent description, highlighting features and analogies, and easing comparisons. Implementation of these methods is also tested on a common edge digital hardware platform to estimate not only abstract functional and complexity features, but also more practical running times and memory footprints on which compliance with real-world applications hinges.

scholarly journals Streaming Algorithms for Subspace Analysis: an Edge- and Hardware-oriented review

2021 ◽  
Author(s):  
Alex Marchioni ◽  
Luciano Prono ◽  
Mauro Mangia ◽  
Fabio Pareschi ◽  
Riccardo Rovatti ◽  
...  
Keyword(s):  
High Dimensional ◽  
Streaming Algorithms ◽  
Subspace Analysis ◽  
Basic Tool ◽  
Hardware Platform ◽  
Common Edge ◽  
Redundant Data ◽  
Real World Applications ◽  
Early Processing ◽  
Digital Hardware

Subspace analysis is a basic tool for coping with high-dimensional data and is becoming a fundamental step in early processing of many signals elaboration tasks. Nowadays trend of collecting huge quantities of usually very redundant data by means of decentralized systems suggests these techniques be deployed as close as possible to the data sources. Regrettably, despite its conceptual simplicity, subspace analysis is ultimately equivalent to eigenspace computation and brings along non-negligible computational and memory requirements. To make this fit into typical systems operating at the edge, specialized streaming algorithms have been recently devised that we here classify and review giving them a coherent description, highlighting features and analogies, and easing comparisons. Implementation of these methods is also tested on a common edge digital hardware platform to estimate not only abstract functional and complexity features, but also more practical running times and memory footprints on which compliance with real-world applications hinges.

scholarly journals Parallel real-world LU decomposition: Gauss vs. Crout algorithm

2018 ◽  
Vol 8 (1) ◽  
pp. 210-217
Author(s):  
Marek Stabrowski
Keyword(s):  
Real World ◽  
Numerical Experiments ◽  
Matrix Decomposition ◽  
Decomposition Algorithms ◽  
Density Matrices ◽  
Hardware Platform ◽  
Posix Threads ◽  
Real World Applications ◽  
Matrix Decomposition Algorithms ◽  
Algorithm Implementation

Abstract This paper presents numerical experiments with assorted versions of parallel LU matrix decomposition algorithms (Gauss and Crout algorithm). The tests have been carried out on the hardware platform with fourcore Skylake processor featuring hyperthreading technology doubling virtually core number. Parallelization algorithms have been implemented with the aid of classic POSIX threads library. Experiments have shown that basic 4-thread acceleration of all parallel implementations is almost equal to the number of threads/processors. Both algorithms are worth considering in real-world applications (Florida University collection). Gauss algorithm is a better performer, with respect to timing, in the case of matrices with lower density of nonzeros, as opposed to higher density matrices. The latter are processed more efficiently with the aid of Crout algorithm implementation.

scholarly journals Highly multiplexed immunofluorescence images and single-cell data of immune markers in tonsil and lung cancer

2019 ◽  
Vol 6 (1) ◽  
Author(s):  
Rumana Rashid ◽  
Giorgio Gaglia ◽  
Yu-An Chen ◽  
Jia-Ren Lin ◽  
Ziming Du ◽  
...  
Keyword(s):  
Lung Cancer ◽  
Single Cell ◽  
High Dimensional ◽  
Cancer Tissue ◽  
Lung Cancer Tissue ◽  
Immune Markers ◽  
Basic Tool ◽  
Human Tonsil ◽  
Biological Discovery ◽  
Cell Data

AbstractIn this data descriptor, we document a dataset of multiplexed immunofluorescence images and derived single-cell measurements of immune lineage and other markers in formaldehyde-fixed and paraffin-embedded (FFPE) human tonsil and lung cancer tissue. We used tissue cyclic immunofluorescence (t-CyCIF) to generate fluorescence images which we artifact corrected using the BaSiC tool, stitched and registered using the ASHLAR algorithm, and segmented using ilastik software and MATLAB. We extracted single-cell features from these images using HistoCAT software. The resulting dataset can be visualized using image browsers and analyzed using high-dimensional, single-cell methods. This dataset is a valuable resource for biological discovery of the immune system in normal and diseased states as well as for the development of multiplexed image analysis and viewing tools.

SOLUTIONS AND DIAGNOSTICS OF SWITCHING PROBLEMS WITH LINEAR STATE DYNAMICS

2016 ◽  
Vol 57 (3) ◽  
pp. 339-351
Author(s):  
J. HINZ ◽  
N. YAP
Keyword(s):  
Optimal Control ◽  
Monte Carlo ◽  
Optimal Control Problems ◽  
Numerical Solutions ◽  
High Dimensional ◽  
Control Problems ◽  
Specific Class ◽  
Real World Applications ◽  
Switching Type ◽  
Linear State

Optimal control problems of stochastic switching type appear frequently when making decisions under uncertainty and are notoriously challenging from a computational viewpoint. Although numerous approaches have been suggested in the literature to tackle them, typical real-world applications are inherently high dimensional and usually drive common algorithms to their computational limits. Furthermore, even when numerical approximations of the optimal strategy are obtained, practitioners must apply time-consuming and unreliable Monte Carlo simulations to assess their quality. In this paper, we show how one can overcome both difficulties for a specific class of discrete-time stochastic control problems. A simple and efficient algorithm which yields approximate numerical solutions is presented and methods to perform diagnostics are provided.

Constraint Solving and Optimization Using Evolutionary Techniques

2019 ◽  
Author(s):  
Mahdi Bidar ◽  
Malek Mouhoub
Keyword(s):  
Optimization Algorithm ◽  
Optimization Problem ◽  
Optimization Problems ◽  
Constraint Solving ◽  
High Dimensional ◽  
Constraint Optimization ◽  
Objective Functions ◽  
Search Spaces ◽  
Real World Applications ◽  
Constraint Optimization Problems

Constraint Solving and Optimization is very relevant in many real world applications including scheduling, planning, configuration, resource allocation and timetabling. Solving a constraint optimization problem consists of finding an assignment of values to variables that optimizes some defined objective functions, subject to a set of constraints imposed on the problem variables. Due to their high dimensional and exponential search spaces, classical methods are unpractical to tackle these problems. An appropriate alternative is to rely on metaheuristics. My thesis is concerned with investigating the applicability of the evolutionary algorithms when dealing with constraint optimization problems. In this regard, we propose two new optimization algorithms namely Mushroom Reproduction Optimization algorithm (MRO) and Focus Group Optimization algorithm (FGO) for solving such problems.

scholarly journals A Hybrid Relational Database Model for Uncertain and Imprecise Information

2020 ◽  
Vol 8 (3) ◽  
Author(s):  
Hoa Nguyen
Keyword(s):  
Fuzzy Set ◽  
Relational Database ◽  
Uncertain Information ◽  
Database Model ◽  
Redundant Data ◽  
Probabilistic Database ◽  
Imprecise Information ◽  
Real World Applications ◽  
Database Models ◽  
Algebraic Operations

Recent years, many fuzzy or probabilistic database models have been built for representing and handling imprecise or uncertain information of objects in real-world applications. However, relational database models combining the relevance and strength of both fuzzy set and probability theories have rarely been proposed. This paper introduces a new relational database model, as a hybrid one combining consistently fuzzy set theory and probability theory for modeling and manipulating uncertain and imprecise information, where the uncertainty and imprecision of a relational attribute value are represented by a fuzzy probabilistic triple, the computation and combination of relational attribute values are implemented by using the probabilistic interpretation of binary relations on fuzzy sets, and the elimination of redundant data is dealt with by coalescing e-equivalent tuples. The basic concepts of the classical relational database model are extended in this new model. Then the relational algebraic operations are formally defined accordingly. A set of the properties of the relational algebraic operations is also formulated and proven.

Digital Hardware Implementation of High Dimensional Fuzzy Systems

2007 ◽  
pp. 245-252 ◽  
Author(s):  
Pablo Echevarria ◽  
M. Victoria Martínez ◽  
Javier Echanobe ◽  
Inés del Campo ◽  
Jose M. Tarela
Keyword(s):  
Fuzzy Systems ◽  
Hardware Implementation ◽  
High Dimensional ◽  
Digital Hardware

scholarly journals High-dimensional data analysis with subspace comparison using matrix visualization

2017 ◽  
Vol 18 (1) ◽  
pp. 94-109 ◽  
Author(s):  
Junpeng Wang ◽  
Xiaotong Liu ◽  
Han-Wei Shen
Keyword(s):  
Dimensionality Reduction ◽  
Dimensional Space ◽  
High Dimensional Data ◽  
Principal Component ◽  
Data Exploration ◽  
High Dimensional ◽  
Data Sets ◽  
Subspace Analysis ◽  
Matrix Visualization ◽  
Fine Tune

Due to the intricate relationship between different dimensions of high-dimensional data, subspace analysis is often conducted to decompose dimensions and give prominence to certain subsets of dimensions, i.e. subspaces. Exploring and comparing subspaces are important to reveal the underlying features of subspaces, as well as to portray the characteristics of individual dimensions. To date, most of the existing high-dimensional data exploration and analysis approaches rely on dimensionality reduction algorithms (e.g. principal component analysis and multi-dimensional scaling) to project high-dimensional data, or their subspaces, to two-dimensional space and employ scatterplots for visualization. However, the dimensionality reduction algorithms are sometimes difficult to fine-tune and scatterplots are not effective for comparative visualization, making subspace comparison hard to perform. In this article, we aggregate high-dimensional data or their subspaces by computing pair-wise distances between all data items and showing the distances with matrix visualizations to present the original high-dimensional data or subspaces. Our approach enables effective visual comparisons among subspaces, which allows users to further investigate the characteristics of individual dimensions by studying their behaviors in similar subspaces. Through subspace comparisons, we identify dominant, similar, and conforming dimensions in different subspace contexts of synthetic and real-world high-dimensional data sets. Additionally, we present a prototype that integrates parallel coordinates plot and matrix visualization for high-dimensional data exploration and incremental dimensionality analysis, which also allows users to further validate the dimension characterization results derived from the subspace comparisons.

Sign in / Sign up

Export Citation Format

Share Document