Dimensions Of Descriptors Research Articles

A comprehensive evaluation of liposome/water partition coefficient prediction models based on the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) method: Challenges from different descriptor dimension reduction methods and machine learning algorithms.

The liposome/water partition coefficient (Klip/w) is a key parameter to evaluate the bioaccumulation potential of pollutants. Considering that it is difficult to determine the Klip/w values of all pollutants through experiments, researchers gradually developed models to predict it. However, there is currently no research on how to comprehensively evaluate prediction models and recommend a compelling optimal modeling method. To remedy the defect of single parameters in a traditional model comparison, the TOPSIS evaluation method, based on entropy weight, was first proposed. We use this method to comprehensively evaluate models from multiple angles in this study. Thirty QSPR models, including 3 descriptor dimension reduction methods and 10 algorithms (belonging to 4 tribes), were used to predict Klip/w and verify the effectiveness of the comprehensive assessment method. The results showed that RF (descriptor dimension reduction method), symbolism (tribes) and RF (algorithm) exhibited significant advantages in establishing the Klip/w value prediction model. At present, the application of TOPSIS in environmental model evaluations is almost absent. We hope that the proposed TOPSIS evaluation method can be applied to more chemical datasets and provide a more systematic and comprehensive basis for the application of the QSPR model in environmental studies and other fields.

Why-So-Deep: Towards Boosting Previously Trained Models for Visual Place Recognition

Deep learning-based image retrieval techniques for the loop closure detection demonstrate satisfactory performance. However, it is still challenging to achieve high-level performance based on previously trained models in different geographical regions. This letter addresses the problem of their deployment with simultaneous localization and mapping (SLAM) systems in the new environment. The general baseline approach uses additional information, such as GPS, sequential keyframes tracking, and re-training the whole environment to enhance the recall rate. We propose a novel approach for improving image retrieval based on previously trained models. We present an intelligent method, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MAQBOOL</i> , to amplify the power of pre-trained models for better image recall and its application to real-time multiagent SLAM systems. We achieve comparable image retrieval results at a low descriptor dimension (512-D), compared to the high descriptor dimension (4096-D) of state-of-the-art methods. We use spatial information to improve the recall rate in image retrieval on pre-trained models. Material related to this work is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://usmanmaqbool.github.io/why-so-deep</uri> .

Machine learning surrogate models for prediction of point defect vibrational entropy

The temperature variation of the defect densities in a crystal depends on vibrational entropy. This contribution to the system thermodynamics remains computationally challenging as it requires a diagonalization of the system's Hessian which scales as O(N3) for a crystal made of N atoms. Here, to circumvent such a heavy computational task and make it feasible even for systems containing millions of atoms, the harmonic vibrational entropy of point defects is estimated directly from the relaxed atomic positions through a linear-in-descriptor machine learning approach of order O(N). With a size-independent descriptor dimension and fixed model parameters, an excellent predictive power is demonstrated on a wide range of defect configurations, supercell sizes, and external deformations well outside the training database. In particular, formation entropies in a range of 250kB are predicted with less than 1.6kB error from a training database whose formation entropies span only 25kB (training error less than 1.0kB). This exceptional transferability is found to hold even when the training is limited to a low-energy superbasin in the phase space while the tests are performed for a different liquid-like superbasin at higher energies.

Improving pattern spotting in historical documents using feature pyramid networks

Pattern spotting consists of locating different instances of a given object (i.e. an image query) in a collection of historical document images. These patterns may vary in shape, size, color, context and even style because they are hand-drawn, which makes pattern spotting a difficult task. To tackle this problem, we propose a Convolutional Neural Network (CNN) approach based on Feature Pyramid Networks (FPN) as the feature extractor of our system. Using FPN allows to extract descriptors of local regions of the documents to be indexed and queries, at multiple scales with just a single forward pass. Experiments conducted on DocExplore dataset show that the proposed system improves mAP by 73% (from 0.157 to 0.272) in pattern localization compared with state-of-the-art results, even when the feature extractor is not trained with domain-specific data. Memory requirement and computation time are also decreased since the descriptor dimension used for distance computation is reduced by a factor of 16.

Defective tissue identification from crowded tissue cluster of 3D images

This research work formulates the challenge of defective tissue identification in crowded tissue clusters. An identified set of 3D tissue images are matched with predefined defective tissue images by using image evidence. Image evidence is obtained by foreground extraction and boundary probability. An optimum solution is derived by the decomposition of conjointly related tissues hooked on occluded plus unoccluded ones in every sequence of iteration. Only unoccluded candidates will be the matching models by the tissue relations of graph description. Tissue authentication and elimination processes are built on minimal descriptor dimension and confined occlusion, which is conceded out after each and every sequence of model equivalent. The computational cost of the proposed method is comparatively much smaller than the comprehensive optimization techniques. Tissue Detection rate is identified to be 2% higher than the result provided by previous means. The performance is demonstrated by the proposed method using another set of challenging tissue datasets. The outcome of our proposed technique is the best than the existing OTSU, HMT, 2DORG, and WK methods.

Classification of alkaloids according to the starting substances of their biosynthetic pathways using graph convolutional neural networks

BackgroundAlkaloids, a class of organic compounds that contain nitrogen bases, are mainly synthesized as secondary metabolites in plants and fungi, and they have a wide range of bioactivities. Although there are thousands of compounds in this class, few of their biosynthesis pathways are fully identified. In this study, we constructed a model to predict their precursors based on a novel kind of neural network called the molecular graph convolutional neural network. Molecular similarity is a crucial metric in the analysis of qualitative structure–activity relationships. However, it is sometimes difficult for current fingerprint representations to emphasize specific features for the target problems efficiently. It is advantageous to allow the model to select the appropriate features according to data-driven decisions for extracting more useful information, which influences a classification or regression problem substantially.ResultsIn this study, we applied a neural network architecture for undirected graph representation of molecules. By encoding a molecule as an abstract graph and applying "convolution" on the graph and training the weight of the neural network framework, the neural network can optimize feature selection for the training problem. By incorporating the effects from adjacent atoms recursively, graph convolutional neural networks can extract the features of latent atoms that represent chemical features of a molecule efficiently. In order to investigate alkaloid biosynthesis, we trained the network to distinguish the precursors of 566 alkaloids, which are almost all of the alkaloids whose biosynthesis pathways are known, and showed that the model could predict starting substances with an averaged accuracy of 97.5%.ConclusionWe have showed that our model can predict more accurately compared to the random forest and general neural network when the variables and fingerprints are not selected, while the performance is comparable when we carefully select 507 variables from 18000 dimensions of descriptors. The prediction of pathways contributes to understanding of alkaloid synthesis mechanisms and the application of graph based neural network models to similar problems in bioinformatics would therefore be beneficial. We applied our model to evaluate the precursors of biosynthesis of 12000 alkaloids found in various organisms and found power-low-like distribution.

Online covariance estimation for novelty‐based visual obstacle detection

AbstractRobust obstacle detection remains a challenge for mobile robots traversing outdoor field environments. Obstacle detection systems that combine multiple cues can potentially overcome deficiencies in individual cues. A key challenge in designing multi‐sensor obstacle detection systems is to automatically and appropriately combine these cues in an unsupervised manner. This paper presents a method for obstacle detection, which continuously adapts its obstacle definition and the weighting of each cue for the current conditions. The key contribution of this paper is a method for online covariance estimation for Parzen windows probability density estimation, which in this application determines the relative importance of each descriptor dimension. By iteratively estimating the covariance using small subsets of the available data, the proposed method is capable of converging to an approximate solution order of magnitude faster than standard optimizers, making it suitable for use online. The application of this covariance estimation method to our novelty‐based obstacle detection system improves obstacle detection precision and reduces learning duration after environmental transitions. It also removes all environment‐specific parameters from the method, and allows the descriptor to contain arbitrary data without time‐consuming hand‐tuning.

Rethinking the sGLOH Descriptor.

sGLOH (shifting GLOH) is a histogram-based keypoint descriptor that can be associated to multiple quantized rotations of the keypoint patch without any recomputation. This property can be exploited to define the best distance between two descriptor vectors, thus avoiding computing the dominant orientation. In addition, sGLOH can reject incongruous correspondences by adding a global constraint on the rotations either as an a priori knowledge or based on the data. This paper thoroughly reconsiders sGLOH and improves it in terms of robustness, speed and descriptor dimension. The revised sGLOH embeds more quantized rotations, thus yielding more correct matches. A novel fast matching scheme is also designed, which significantly reduces both computation time and memory usage. In addition, a new binarization technique based on comparisons inside each descriptor histogram is defined, yielding a more compact, faster, yet robust alternative. Results on an exhaustive comparative experimental evaluation show that the revised sGLOH descriptor incorporating the above ideas and combining them according to task requirements, improves in most cases the state of the art in both image matching and object recognition.

Low Dimensional Representation of Fisher Vectors for Microscopy Image Classification.

Microscopy image classification is important in various biomedical applications, such as cancer subtype identification, and protein localization for high content screening. To achieve automated and effective microscopy image classification, the representative and discriminative capability of image feature descriptors is essential. To this end, in this paper, we propose a new feature representation algorithm to facilitate automated microscopy image classification. In particular, we incorporate Fisher vector (FV) encoding with multiple types of local features that are handcrafted or learned, and we design a separation-guided dimension reduction method to reduce the descriptor dimension while increasing its discriminative capability. Our method is evaluated on four publicly available microscopy image data sets of different imaging types and applications, including the UCSB breast cancer data set, MICCAI 2015 CBTC challenge data set, and IICBU malignant lymphoma, and RNAi data sets. Our experimental results demonstrate the advantage of the proposed low-dimensional FV representation, showing consistent performance improvement over the existing state of the art and the commonly used dimension reduction techniques.

Settling velocity of non-spherical hydrochorous seeds

The settling velocity of irregular particles is affected by both particle properties (density, size, and shape) and fluid properties (density and viscosity) and is therefore a function of particle Reynolds number and particle shape. Experimental data from the literature have been collected on a wide range of particle Reynolds numbers (0.03–10,000) to develop empirical formulae for drag coefficients and settling velocities of non-spherical particles. Three shape factors were adopted and different dimensions of descriptors were recommended according to the types of particles. For coarse particles with large dimensionless diameter, the effect of shape on settling velocity is more significant, and thus piecewise formulae were derived to further improve prediction accuracy. Compared with other models, the newly presented formulae predict more accurately the settling velocity of irregular particles. Moreover, settling velocities of three types of hydrochorous seeds collected from Poyang Lake were measured in the laboratory and the comparison between the measured and predicted settling velocity shows that the empirical formulae can be applied to non-spherical granular hydrochorous seeds and can enrich the study on the dynamic pattern of hydrochory.

Learning a non‐linear combination of Mahalanobis distances using statistical inference for similarity measure

In this study, the authors learn a similarity measure that discriminates between inter‐class and intra‐class samples based on a statistical inference perspective. A non‐linear combination of Mahalanobis is proposed to reflect the properties of a likelihood ratio test. Since an object's appearance is influenced by the identity of the object and variations in the capturing process, the authors represent the feature vector, which is the difference between two samples in the differences space, as a sample that is drawn from a mixture of many distributions. This mixture consists of the identities distribution and other distributions of the variations in the capturing process, in case of dissimilar samples. However, in the case of similar samples, the mixture consists of the variations in the capturing process distributions only. Using this representation, the proposed similarity measure accurately discriminates between inter‐class and intra‐class samples. To highlight the good performance of the proposed similarity measure, it is tested on different computer vision applications: face verification and person re‐identification. To illustrate how the proposed learning method can easily be used on large scale datasets, experiments are conducted on different challenging datasets: labelled faces in the wild (LFW), public figures face database, ETHZ and VIPeR. Moreover, in these experiments, the authors evaluate different stages, for example, features detector, descriptor type and descriptor dimension, which constitute the face verification pipeline. The experimental results confirm that the learning method outperforms the state‐of‐the‐art.

Notes on quantitative structure‐properties relationships (QSPR) (1): A discussion on a QSPR dimensionality paradox (QSPR DP) and its quantum resolution

Classical quantitative structure-properties relationship (QSPR) statistical techniques unavoidably present an inherent paradoxical computational context. They rely on the definition of a Gram matrix in descriptor spaces, which is used afterwards to reduce the original dimension via several possible kinds of algebraic manipulations. From there, effective models for the computation of unknown properties of known molecular structures are obtained. However, the reduced descriptor dimension causes linear dependence within the set of discrete vector molecular representations, leading to positive semi-definite Gram matrices in molecular spaces. To resolve this QSPR dimensionality paradox (QSPR DP) here is proposed to adopt as starting point the quantum QSPR (QQSPR) computational framework perspective, where density functions act as infinite dimensional descriptors. The fundamental QQSPR equation, deduced from employing quantum expectation value numerical evaluation, can be approximately solved in order to obtain models exempt of the QSPR DP. The substitution of the quantum similarity matrix by an empirical Gram matrix in molecular spaces, build up with the original non manipulated discrete molecular descriptor vectors, permits to obtain classical QSPR models with the same characteristics as in QQSPR, that is: possessing a certain degree of causality and explicitly independent of the descriptor dimension.

Prediction of selectivity index of pentachlorophenol-imprinted polymers

A data set comprising of the selectivity index of pentachlorophenol-imprinted polymers against 53 pentachlorophenol and related compounds was obtained from the excellent work of Baggiani et al. Molecular descriptors of the phenol compounds were calculated with EDRAGON to obtain a total of 1,666 descriptors spanning 20 categories of molecular properties. Multivariate analysis of the data set was performed using multiple linear regression, partial least squares regression, and principal component regression. Partial least squares regression was found to deliver an excellent predictive model and was chosen for further investigation. The descriptor dimension was reduced by the combined use of partial least squares and Unsupervised Forward Selection algorithm. The obtained Quantitative Structure-Property Relationship (QSPR) model based on the smaller subset of the molecular descriptors displayed substantial gain in predictive ability when compared to the model of Baggiani et al. Such QSPR model can help in the computational design of MIPs with predefined selectivity toward template molecules of interest.

A multi-dimensional classification of production systems for the design and selection of production planning and control systems

A primary requirement for improved understanding of the management of production systems is an appropriate classification of such systems. This paper proposes a classification that facilitates a better understanding of real production systems. It combines all the essential features, e.g. the flow of materials with new classification perspectives with respect to response time, repetitiveness and work organization. As much previous work has influenced the approach proposed here, the paper also presents a review of relevant literature. The classification has four groups of characteristics, comprising eight dimensions of descriptors, encompassing 12 variables. Choosing or designing an appropriate production planning and control system (PPCS) is a difficult task due to the integrative character of the production planning and control area: it tends to interface with all functional areas of the enterprise. An important objective of the classification system proposed here is to provide a tool to assist in undertaking this difficult task. Real production systems are becoming more hybrid in order to be able to cope with change. We show that our classification can successfully deal with such systems.

Principal dimensions of the pain experience and psychological disturbance in chronic low back pain patients

The principal dimensions of the pain experience of 102 patients with chronic low back pain were studied and components of the pain experience were compared to pre-treatment measures of emotional disturbance. Responses on the McGill Pain Questionnaire (MPQ) were factor analyzed revealing dimensions of the pain experience that seemed to represent sensory, affective, and evaluative aspects. Patients with high scores on the MMPI hypochondriasis scale portrayed their pain as more intense and as high in terms of affective and evaluative descriptors on the MPQ. Other MMPI scales reflecting emotional disturbance (depression, hysteria) were also associated with a more intense description of pain and higher scores on an affective descriptor dimension. The evaluative descriptive dimension was highly related to the overall intensity of the pain experience. Only the affective dimension was related to emotional disturbance separately from the effects of intensity alone.