Deep learning model with low-dimensional random projection for large-scale image search

Alzu’bi, Ahmad; Abuarqoub, Abdelrahman

doi:10.1016/j.jestch.2019.12.004

Cited by 11 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of applications, variants of the RP algorithm have been successfully applied to address some of the most important challenges of big data systems, including privacy protection [27,28], handling of high-dimensional data [6,29], and system scalability [7,30,31], among many others.…”

Section: Random Projection Variantsmentioning

confidence: 99%

“…Thanks to this property, Random Projection has become a widespread tool for dimensionality reduction, especially in large-scale applications where the volume of data or the dimensionality of samples is too big for alternative methods. For instance, Random Projection has been successfully used to accelerate tasks such as multivariate correlation analysis [4], high-dimensional data clustering [5,6], image search [7] or texture classification [8], among many others.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Tuning Database-Friendly Random Projection Matrices for Improved Distance Preservation on Specific Data

López-Sánchez

Bodt

Lee³

et al. 2021

Appl Intell

View full text Add to dashboard Cite

Random Projection is one of the most popular and successful dimensionality reduction algorithms for large volumes of data. However, given its stochastic nature, different initializations of the projection matrix can lead to very different levels of performance. This paper presents a guided random search algorithm to mitigate this problem. The proposed method uses a small number of training data samples to iteratively adjust a projection matrix, improving its performance on similarly distributed data. Experimental results show that projection matrices generated with the proposed method result in a better preservation of distances between data samples. Conveniently, this is achieved while preserving the database-friendliness of the projection matrix, as it remains sparse and comprised exclusively of integers after being tuned with our algorithm. Moreover, running the proposed algorithm on a consumer-grade CPU requires only a few seconds.

show abstract

Section: Random Projection Variantsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Tuning Database-Friendly Random Projection Matrices for Improved Distance Preservation on Specific Data

López-Sánchez

Bodt

Lee³

et al. 2021

Appl Intell

View full text Add to dashboard Cite

show abstract

“…In LSTM structure, both x t and ℎ t-1 used as inputs and which data should be deleted are firstly determined. [31][32] This process is performed in the forget layer (f t ) with the equations shown in Table 2. Also, the components of i, f, g, and o are calculated with these equations.…”

Section: The Construction Of the Lstm Modelmentioning

confidence: 99%

“…The error concept in the prediction can be defined as the differences between the predicted and actual values. [30][31][32] To evaluate the performance of the proposed model, some metrics are used as mean absolute error T A B L E 2 Functions used in LSTM structure…”

Section: The Performance Of the Proposed Dnn Modelmentioning

confidence: 99%

“…The error concept in the prediction can be defined as the differences between the predicted and actual values 30‐32 . To evaluate the performance of the proposed model, some metrics are used as mean absolute error (MAE), root mean squared error (RMSE), mean absolute percentage error (MAPE), and R ‐squared ( R 2 ) that are given in Equations ).

MAE = \frac{1}{n} {falsefalse}_{\sum}^{i = 1} (||, y_{italicmeas_i} - y_{italicmodel_i})

RMSE = \sqrt{\frac{\sum_{i = 1}^{n} {(y_{italicmeas_i} - y_{italicmodel_i})}^{2}}{n}}

MAPE = \frac{100 %}{n} \sum_{i = 1}^{n} \frac{|y_{meas_i} - y_{model_i}|}{|y_{meas_i}|}

R^{2} = 1 - \frac{\sum_{i = 1}^{n} {(y_{mea s_{i}} - y_{mode l_{i}})}^{2}}{\sum_{i = 1}^{n} {(y_{mea s_{i}} - {\overset{y}{true}}_{italicmeas})}^{2}}

where y meas_i is the measured value at i th row, y model_i is the predicted value at i th row, and

{\overset{false¯}{y}}_{meas}

is the average of all predicted ones.…”

Section: The Design Of Dnn Modelmentioning

confidence: 99%

See 1 more Smart Citation

The dielectric properties prediction of the vegetation depending on the moisture content using the deep neural network model

Metlek

Kayaalp

Başyiğit

et al. 2020

Int J RF Microw Comput Aided Eng

View full text Add to dashboard Cite

In this paper, dielectric properties of citrus leaves are predicted with long shortterm memory (LSTM) which is one of the well-known deep neural network (DNN) models and real-time measurements for any moisture content (MC) values in the range of 4.90 to 7.05 GHz at a fixed temperature of 24 C for microwave applications, as a novelty. Firstly, S-parameters of samples are measured with WR-159 waveguide and Waveguide Transmission Line Method. In addition, the MCs of samples depending on their weights are calculated. Thus, the dataset depending on various MC and frequency is obtained with the measurement results to both training and testing the DNN model. Secondly, a total of 4000 datasets are obtained, 80% of which is used for training, and 20% for testing. The proposed DNN model consists of four inputs (f, MC, S 11 , and S 21) and two outputs (ε 0 and ε 00). Finally, the dielectric parameters for the desired MC and f are displayed with the graphical user interface in real-time. Success criteria for the prediction such as mean absolute error, root mean squared error, mean absolute percentage error, and R-squared are calculated. The results indicated that there is good agreement between the measured and predicted ones. R-squared are calculated as 0.962 and 0.968 for ε 0 and ε 00 , respectively.

show abstract

New algorithms for trace-ratio problem with application to high-dimension and large-sample data dimensionality reduction

Shi

2021

Mach Learn

View full text Add to dashboard Cite

Learning large-scale data sets with high dimensionality is a main concern in research areas including machine learning, visual recognition, information retrieval, to name a few. In many practical uses such as images, video, audio, and text processing, we have to face with high-dimension and large-sample data problems. The trace-ratio problem is a key problem for feature extraction and dimensionality reduction to circumvent the high dimensional space. However, it has been long believed that this problem has no closed-form solution, and one has to solve it by using some inner-outer iterative algorithms that are very time consuming. Therefore, efficient algorithms for high-dimension and large-sample trace-ratio problems are still lacking, especially for dense data problems. In this work, we present a closed-form solution for the trace-ratio problem, and propose two algorithms to solve it. Based on the formula and the randomized singular value decomposition, we first propose a randomized algorithm for solving high-dimension and large-sample dense traceratio problems. For high-dimension and large-sample sparse trace-ratio problems, we then propose an algorithm based on the closed-form solution and solving some consistent under-determined linear systems. Theoretical results are established to show the rationality and efficiency of the proposed methods. Numerical experiments are performed on some real-world data sets, which illustrate the superiority of the proposed algorithms over many state-of-the-art algorithms for high-dimension and large-sample dimensionality reduction problems. KeywordsDimensionality reduction • Trace-ratio problem • High-dimension and large-sample data • Large-scale discriminant analysis • Randomized singular value decomposition (RSVD) • Inner-outer iterative algorithm Editors: Tim Verdonck, Bart Baesens, María Óskarsdóttir and Seppe vanden Broucke.

show abstract

Deep learning model with low-dimensional random projection for large-scale image search

Cited by 11 publications

References 11 publications

Tuning Database-Friendly Random Projection Matrices for Improved Distance Preservation on Specific Data

Tuning Database-Friendly Random Projection Matrices for Improved Distance Preservation on Specific Data

The dielectric properties prediction of the vegetation depending on the moisture content using the deep neural network model

New algorithms for trace-ratio problem with application to high-dimension and large-sample data dimensionality reduction

Contact Info

Product

Resources

About