Search

Images, texts, voices, and signals can be synthesized by latent spaces in a multidimensional vector, which can be explored without the hurdles of noise or other interfering factors. In this paper, we present a practical use case that demonstrates the power of latent space in exploring complex realities such as image space. We focus on DaVinciFace, an AI-based system that explores the StyleGAN2 space to create a high-quality portrait for anyone in the style of the Renaissance genius Leonardo da Vinci. The user enters one of their portraits and receives the corresponding Da Vinci-style portrait as an output. Since most of Da Vinci's artworks depict young and beautiful women (e.g., "La Belle Ferroniere", "Beatrice de' Benci"), we investigate the ability of DaVinciFace to account for other social categorizations, including gender, race, and age. The experimental results evaluate the effectiveness of our methodology on 1158 portraits acting on the vector representations of the latent space to produce high-quality portraits that retain the facial features of the subject's social categories, and conclude that sparser vectors have a greater effect on these features. To objectively evaluate and quantify our results, we solicited human feedback via a crowd-sourcing campaign. Analysis of the human feedback showed a high tolerance for the loss of important identity features in the resulting portraits when the Da Vinci style is more pronounced, with some exceptions, including Africanized individuals. [ABSTRACT FROM AUTHOR]

To address the nonlinearity and dynamics of industrial processes, an incremental isometric mapping (IISOMAP) in combination with double local density (DLD) is proposed as a fault detection method (IISOMAP-DLD) based on stream shape learning. First, IISOMAP is used to map the raw data into a low-dimensional manifold feature subspace and a residual subspace. Then, the double local density method is introduced in the two subspaces respectively to construct statistics to monitor the process. Finally, the IISOMAP-DLD method is applied to the Tennessee-Eastman (TE) process, and the experimental results show that IISOMAP-DLD has a higher fault detection rate than the other methods. IISOMAP preserves the intrinsic characteristics of the data and solves the nonlinear problems of the process, while the double local density method can eliminate the dynamic of the process.

Copyright of Journal of Shanghai Jiao Tong University (1006-2467) is the property of Journal of Shanghai Jiao Tong University Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Images, texts, voices, and signals can be synthesized by latent spaces in a multidimensional vector, which can be explored without the hurdles of noise or other interfering factors. In this paper, we present a practical use case that demonstrates the power of latent space in exploring complex realities such as image space. We focus on DaVinciFace, an AI-based system that explores the StyleGAN2 space to create a high-quality portrait for anyone in the style of the Renaissance genius Leonardo da Vinci. The user enters one of their portraits and receives the corresponding Da Vinci-style portrait as an output. Since most of Da Vinci’s artworks depict young and beautiful women (e.g., “La Belle Ferroniere”, “Beatrice de’ Benci”), we investigate the ability of DaVinciFace to account for other social categorizations, including gender, race, and age. The experimental results evaluate the effectiveness of our methodology on 1158 portraits acting on the vector representations of the latent space to produce high-quality portraits that retain the facial features of the subject’s social categories, and conclude that sparser vectors have a greater effect on these features. To objectively evaluate and quantify our results, we solicited human feedback via a crowd-sourcing campaign. Analysis of the human feedback showed a high tolerance for the loss of important identity features in the resulting portraits when the Da Vinci style is more pronounced, with some exceptions, including Africanized individuals.

Predictive analytics of financial markets in developed and emerging economies during the COVID-19 regime is undeniably challenging due to unavoidable uncertainty and the profound proliferation of negative news on different platforms. Tracking the media echo is crucial to explaining and anticipating the abrupt fluctuations in financial markets. The present research attempts to propound a robust framework capable of channeling macroeconomic reflectors and essential media chatter-linked variables to draw precise forecasts of future figures for Spanish and Indian stock markets. The predictive structure combines Isometric Mapping (ISOMAP), which is a non-linear feature transformation tool, and Gradient Boosting Regression (GBR), which is an ensemble machine learning technique to perform predictive modelling. The Explainable Artificial Intelligence (XAI) is used to interpret the black-box type predictive model to infer meaningful insights. The overall results duly justify the incorporation of local and global media chatter indices in explaining the dynamics of respective financial markets. The findings imply marginally better predictability of Indian stock markets than their Spanish counterparts. The current work strives to compare and contrast the reaction of developed and developing financial markets during the COVID-19 pandemic, which has been argued to share a close resemblance to the Black Swan event when applying a robust research framework. The insights linked to the dependence of stock markets on macroeconomic indicators can be leveraged for policy formulations for augmenting household finance. [ABSTRACT FROM AUTHOR]

The class labels of collected coal and gas outbursts sample data may be wrong, if these collected sample data are directly used for outbursts classification, the accuracy and efficiency are very low. In this paper, a novel hybrid model that integrates fuzzy c-means clustering based on isometric mapping feature extraction and Bayesian optimized mixed kernel relevance vector machine classifier is proposed and applied to improve the classification performance of coal and gas outbursts. First, the isometric mapping is used to extract the non-linear information, then the significant features are selected, in order to improve the classification performance of coal and gas outbursts, the fuzzy c-means is used to perform clustering analysis on the effective features obtained from isometric mapping, mining the structural information and internal regularity of the sample data and estimating classification labels of sample data. Second, A mixed kernel relevance vector machine classifier is proposed to classify coal and gas outbursts, improving the learning and generalization ability of outbursts classification, and the classifier parameters are optimized by Bayesian optimization with global and local search capability remarkably. Finally, the improved fuzzy c-means clustering is integrated into the mixed kernel relevance vector machine classifier model, and Bayesian optimization algorithm is used to help train a better classifier for outbursts classification. The obtained experimental results on the collected actual dataset of coal and gas outbursts show that proposed clustering method can improve the clustering effect and efficiency, decrease the feature vector size up to 50% and achieves the accuracy and running time of 100% and 5.26 s, respectively, which outperforms prior methods with 98% and 5.38 s,the proposed outbursts combined classification model based on classification and clustering model outperforms other methods by 4%-6% with respect to average accuracy. It is believed that the proposed model is very effective for outbursts classification. [ABSTRACT FROM AUTHOR]

Predictive analytics of financial markets in developed and emerging economies during the COVID-19 regime is undeniably challenging due to unavoidable uncertainty and the profound proliferation of negative news on different platforms. Tracking the media echo is crucial to explaining and anticipating the abrupt fluctuations in financial markets. The present research attempts to propound a robust framework capable of channeling macroeconomic reflectors and essential media chatter-linked variables to draw precise forecasts of future figures for Spanish and Indian stock markets. The predictive structure combines Isometric Mapping (ISOMAP), which is a non-linear feature transformation tool, and Gradient Boosting Regression (GBR), which is an ensemble machine learning technique to perform predictive modelling. The Explainable Artificial Intelligence (XAI) is used to interpret the black-box type predictive model to infer meaningful insights. The overall results duly justify the incorporation of local and global media chatter indices in explaining the dynamics of respective financial markets. The findings imply marginally better predictability of Indian stock markets than their Spanish counterparts. The current work strives to compare and contrast the reaction of developed and developing financial markets during the COVID-19 pandemic, which has been argued to share a close resemblance to the Black Swan event when applying a robust research framework. The insights linked to the dependence of stock markets on macroeconomic indicators can be leveraged for policy formulations for augmenting household finance.

The isometric mapping (Isomap) algorithm is often used for analysing hyperspectral images. Isomap allows to reduce such hyperspectral images from a high-dimensional space into a lower-dimensional space, keeping the critical original information. To achieve such objective, Isomap uses the state-of-the-art MultiDimensional Scaling method (MDS) for dimensionality reduction. In this work, we propose to use Isomap with SMACOF, since SMACOF is the most accurate MDS method. A deep comparison, in terms of accuracy, between Isomap based on an eigendecomposition process and Isomap based on SMACOF has been carried out using three benchmark hyperspectral images.Moreover, for the hyperspectral image classification, three classifiers (support vector machine, k-nearest neighbour, and Random Forest) have been used to compare both Isomap approaches. The experimental investigation has shown that better classification accuracy is obtained by Isomap with SMACOF. [ABSTRACT FROM AUTHOR]

Manifold learning algorithms such as the isometric mapping (ISOMAP) algorithm have been widely used in the analysis of hyperspectral images (HSIs), for both visualization and dimension reduction. As advanced versions of the traditional linear projection techniques, the manifold learning algorithms find the low-dimensional feature representation by nonlinear mapping, which can better preserve the local structure of the original data and thus benefit the data analysis. However, the high computational complexity of the manifold learning algorithms hinders their application in HSI processing. Although there are a few parallel implementations of manifold learning approaches that are available in the remote sensing community, they have not been designed to accelerate the eigen-decomposition process, which is actually the most time-consuming part of the manifold learning algorithms. In this letter, as a case study, we discuss the graphics processing unit parallel implementation of the ISOMAP algorithm. In particular, we focus on the eigen-decomposition process and verify the applicability of the proposed method by validating the embedding vectors and the subsequent classification accuracies. The experimental results obtained on different HSI data sets show an excellent speedup performance and consistent classification accuracy compared with the serial implementation. [ABSTRACT FROM PUBLISHER]

Aiming at the poor generalization of the ISOMAP algorithm when applied on new samples, this paper proposed MRC-ISOMAP (manifold reconstruction-ISOMAP) algorithm to reduce sample characteristics' dimension on the basis of having an in-depth understanding of LLE algorithm's local reconstruction idea. For training samples, MRC-ISOMAP adopted ISOMAP algorithm's global nonlinear structure theory to calculate training samples' low-dimensional representations, and for added samples, the algorithm adopted the theory of local linear to maintain local linear relationships which could conduct the new samples' low-dimensional representation reconstructed by low-dimensional training samples more quickly and accurately. Comparing with the original ISOMAP algorithm, experiments on USTB3 ear image library show that MRC-ISOMAP algorithm can get a higher recognition rate, and achieve a higher efficiency when dealing with added samples. [ABSTRACT FROM AUTHOR]

Node localization is a main application of wireless sensor networks. However, the measurement error and noise in a real environment make the positioning of information nonlinear and have a great effect on the performance of localization methods. In this paper, we propose a novel isometric mapping (Isomap) node localization algorithm based on partial least squares (PLS-Isomap). For topological stability, the critical outlier points are eliminated by comparing the contribution rate of all data points. Then, we employ the PLS method to solve the Isomap. The adoption of PLS reduces the noise sensitivity of Isomap, which achieves solution by least squares. Moreover, the proposed approach applies a projection method to construct a new kernel matrix between new and original data points. Compared with Isomap and the multidimensional scale method, experimental and simulation results indicate that the PLS-Isomap algorithm has good topological stability, robustness, positioning accuracy, and lower computational complexity. [ABSTRACT FROM PUBLISHER]

In recent publications, we have presented a data-driven approach to representing the nonlinear structure of hyperspectral imagery using manifold coordinates. The approach relies on graph methods to derive geodesic distances on the high-dimensional hyperspectral data manifold. From these distances, a set of intrinsic manifold coordinates that parameterizes the data manifold is derived. Scaling the solution relied on divide-conquer-and-merge strategies for the manifold coordinates because of the computational and memory scaling of the geodesic coordinate calculations. In this paper, we improve the scaling performance of isometric mapping (ISOMAP) and achieve full-scene global manifold coordinates while removing artifacts generated by the original methods. The CPU time of the enhanced ISOMAP approach scales as O(N log²(N)), where N is the number of samples, while the memory requirement is bounded by O(N log(N)). Full hyperspectral scenes of 0(106) samples or greater are obtained via a reconstruction algorithm, which allows insertion of large numbers of samples into a representative ‘backbone’ manifold obtained for a smaller but representative set of O(105) samples. We provide a classification example using a coastal hyperspectral scene to illustrate the approach. [ABSTRACT FROM AUTHOR]