Minimal Loss Of Information Research Articles

A Novel Weld-Seam Defect Detection Algorithm Based on the S-YOLO Model

Detecting small targets and handling target occlusion and overlap are critical challenges in weld defect detection. In this paper, we propose the S-YOLO model, a novel weld defect detection method based on the YOLOv8-nano model and several mathematical techniques, specifically tailored to address these issues. Our approach includes several key contributions. Firstly, we introduce omni-dimensional dynamic convolution, which is sensitive to small targets, for improved feature extraction. Secondly, the NAM attention mechanism enhances feature representation in the region of interest. NAM computes the channel-wise and spatial-wise attention weights by matrix multiplications and element-wise operations, and then applies them to the feature maps. Additionally, we replace the SPPF module with a context augmentation module to improve feature map resolution and quality. To minimize information loss, we utilize Carafe upsampling instead of the conventional upsampling operations. Furthermore, we use a loss function that combines IoU, binary cross-entropy, and focal loss to improve bounding box regression and object classification. We use stochastic gradient descent (SGD) with momentum and weight decay to update the parameters of our model. Through rigorous experimental validation, our S-YOLO model demonstrates outstanding accuracy and efficiency in weld defect detection. It effectively tackles the challenges of small target detection, target occlusion, and target overlap. Notably, the proposed model achieves an impressive 8.9% improvement in mean Average Precision (mAP) compared to the native model.

A New Approach for Anonymizing Transaction Data with Set Values

This article proposes a new method that can guarantee strong privacy while minimizing information loss in transactional data composed of a set of each attribute value in a relational database, which is not generally well-known structured data. The proposed scheme adopts the same top-down partitioning algorithm as the existing k-anonymity model, using local generalization to optimize safety and CPU execution time. At the same time, the information loss rate, which is a disadvantage of the existing local generalization, is further improved by reallocating transactions through an additional bottom-up tree search process after the partitioning process. Our scheme shows a very fast processing time compared to the HgHs algorithm using generalization and deletion techniques. In terms of information loss, our scheme shows much better performance than any schemes proposed so far, such as the existing local generalization or HgHs algorithm. In order to evaluate the efficiency of our algorithm, the experiment compared its performance with the existing local generalization and the HgHs algorithm, in terms of both execution time and information loss rate. As a result of the experiment, for example, when k is 5 in k-anonymity for the dataset BMS-WebView-2, the execution time of our scheme is up to 255 times faster than the HgHs algorithm, and with regard to the information loss rate, our method showed a maximum rate of 62.37 times lower than the local generalization algorithm.

Generating multi-level explanations for process outcome predictions

Process mining focuses on the analysis of event log data to build various process analytical capabilities. Predictive process analytics has emerged as one of such key capabilities and it uses machine learning techniques to construct process prediction models. In recent years, deep neural networks have gained increasing interest in process prediction since they can handle multi-dimensional sequential inputs with minimal information loss. However, they are considered black-box models and existing studies in explaining deep neural network-based process predictions rely on only event-level features for explanation. In this paper, we propose a new approach for generating explanations for process outcome predictions at multiple levels. The approach is underpinned by three different prediction models: a transparent model for generating global explanations based on case-level features, an attention-based deep neural network for generating local explanations based on event-level features, and a novel eXplainable Dual-learning Deep network (XD2-net) for generating local explanations based on case-level features. Using three publicly available datasets, we have tested the applicability of the approach and further examined the multi-level explanations generated by the approach through an elaborate case study. Unlike others, the design of our approach promotes the idea of leveraging the complementary capabilities of different models and utilizing their strengths, rather than focusing on model performance competition. This will contribute towards generating more comprehensive explanations that meet the needs of different end users and purposes in the future.

Accelerating Open Modification Spectral Library Searching on Tensor Core in High-dimensional Space.

Driven by technological advances, the throughput and cost of mass spectrometry proteomics experiments have improved by orders of magnitude in recent decades. Spectral library searching is a common approach to annotating experimental mass spectra by matching them against large libraries of reference spectra corresponding to known peptides. An important disadvantage, however, is that only peptides included in the spectral library can be found, whereas novel peptides, such as those with unexpected post-translational modifications, will remain unknown. Open modification searching is an increasingly popular approach to annotate modified peptides based on partial matches against their unmodified counterparts. Unfortunately, this leads to very large search spaces and excessive runtimes, which is especially problematic considering the continuously increasing sizes of mass spectrometry proteomics datasets. We propose an open modification searching algorithm, called HOMS-TC, that fully exploits parallelism in the entire pipeline of spectral library searching. We designed a new highly parallel encoding method based on the principle of hyperdimensional computing to encode mass spectral data to hypervectors while minimizing information loss. This process can be easily parallelized since each dimension is calculated independently. HOMS-TC processes two stages of existing cascade search in parallel and selects the most similar spectra while considering post-translational modifications. We accelerate HOMS-TC on NVIDIA's Tensor Core Units, which is emerging and readily available in the recent graphics processing unit (GPU). Our evaluation shows that HOMS-TC is 31 × faster on average than alternative search engines and provides comparable accuracy to competing search tools. HOMS-TC is freely available under the Apache 2.0 license as an open-source software project at https://github.com/tycheyoung/homs-tc.

Object of Interest and Unsupervised Learning-based Framework for an Effective Video Summarization Using Deep Learning

During this digital era, a large amount of visual data is being generated by various multimedia sources. A technique is urgently required to provide a clear and accurate summary of the original video to highlight the most informative segments of the video content. However, the varying video resolution, multi-dimensional feature representation, and massive storage create difficulties in key frame extraction techniques. As a result, many unnecessary features of the videos must be dropped to count their unique qualities. Therefore, a novel deep learning-based approach is proposed where frames are first retrieved using 25 frames per second; then, objects are detected on extracted frames using YOLOv5, and the frames with the target object only are processed further to overcome time consumption and high-speed computing hardware limitations. Further features are obtained using VGG-16 and object of Interest (OoI) based ResNet-50, respectively, and comparisons are performed to find the best solution. The extracted features are compressed using Principal Component Analysis (PCA) based on unsupervised Learning, which may efficiently minimize information loss, and potentially reduce dimension. By performing a comprehensive evaluation to obtain the best value of K using the Silhouette score, Candidate frames are extracted with the maximum mean and standard deviation from each K-means algorithm-based cluster. Pearson Correlation Coefficient (PCC) is used as a post- processing step to remove the redundant frames from the candidate frames and final keyframes extraction. The experiment was performed on the benchmark office dataset from industrial surveillance, which outperforms the state-of-the-art models in terms of recall score.

Atomic Resolution Cryogenic 4D-STEM Imaging via Robust Distortion Correction.

Cryogenic four-dimensional scanning transmission electron microscopy (4D-STEM) imaging is a useful technique for studying quantum materials and their interfaces by simultaneously probing charge, lattice, spin, and chemistry on the atomic scale with the sample held at temperatures ranging from room to cryogenic. However, its applications are currently limited by the instabilities of cryo-stages and electronics. To overcome this challenge, we develop an algorithm to effectively correct the complex distortions present in atomic resolution cryogenic 4D-STEM data sets. This method uses nonrigid registration to identify localized distortions in a 4D-STEM and relate them to an undistorted experimental STEM image, followed by a series of affine transformations for distortion corrections. This method allows a minimum loss of information in both reciprocal and real spaces, enabling the reconstruction of sample information from 4D-STEM data sets. This method is computationally cheap, fast, and applicable for on-the-fly data analysis in future in situ cryogenic 4D-STEM experiments.

Potato Blight Classification Android Application using Deep learning

Farmers who grow potatoes suffer from significant financial losses each year due several diseases that affect potato plants. The most common diseases are Early and Late Blight , caused by fungus and specific microorganisms, respectively. Early detection and appropriate treatment can save a lot of waste and prevent economic losses. However, traditional visual inspection methods are time-consuming and prone to errors. To address this challenge, we propose a Convolutional Neural Network ( CNN ) approach for plant disease diagnosis. CNNs are a type of deep learning algorithm widely used for image classification tasks. They can automatically learn features from input image data, making them well-suited for plant disease diagnosis. Our customized CNN has fewer trainable parameters, reducing computation time and minimizing information loss. We used several convolutional and pooling layers, followed by fully connected layers with the ReLU activation function. We also applied dropout regularization to prevent overfitting. In conclusion, accurate and efficient plant disease diagnosis is essential for preventing economic losses. Our customized CNN for plant disease diagnosis has the potential to be an effective tool for farmers. It can help them diagnose plant diseases quickly and accurately, leading to timely treatment and reduced loss.

Can the Gravitational Effect of Planet X be Detected in Current-era Tracking of the Known Major and Minor Planets?

Using Fisher information matrices, we forecast the uncertainties σ M on the measurement of a “Planet X” at heliocentric distance d X via its tidal gravitational field’s action on the known planets. Using planetary measurements currently in hand, including ranging from the Juno, Cassini, and Mars-orbiting spacecraft, we forecast a median uncertainty (over all sky positions) of A 5σ detection of a 5 M ⊕ Planet X at d X = 400 au should be possible over the full sky but over only 5% of the sky at d X = 800 au. The gravity of an undiscovered Earth- or Mars-mass object should be detectable over 90% of the sky to a distance of 260 or 120 au, respectively. Upcoming Mars ranging improves these limits only slightly. We also investigate the power of high-precision astrometry of ≈8000 Jovian Trojans over the 2023–2035 period from the upcoming Legacy Survey of Space and Time (LSST). We find that the dominant systematic errors in optical Trojan astrometry (photocenter motion, nongravitational forces, and differential chromatic refraction) can be solved internally with minimal loss of information. The Trojan data allow cross-checks with Juno/Cassini/Mars ranging, but do not significantly improve the best achievable σ M values until they are ≳10× more accurate than expected from LSST. The ultimate limiting factor in searches for a Planet X tidal field is confusion with the tidal field created by the fluctuating quadrupole moment of the Kuiper Belt as its members orbit. This background will not, however, become the dominant source of uncertainty until the data get substantially better than they are today.

Developments of Efficient Trigonometric Quantile Regression Models for Bounded Response Data

The choice of an appropriate regression model for econometric modeling minimizes information loss and also leads to sound inferences. In this study, we develop four quantile regression models based on trigonometric extensions of the unit generalized half-normal distributions for the modeling of a bounded response variable defined on the unit interval. The desirable shapes of these distributions, such as left-skewed, right-skewed, reversed-J, approximately symmetric, and bathtub shapes, make them competitive models for bounded responses with such traits. The maximum likelihood method is used to estimate the parameters of the regression models, and Monte Carlo simulation results confirm the efficiency of the method. We demonstrate the utility of our models by investigating the relationship between OECD countries’ educational attainment levels, labor market insecurity, and homicide rates. The diagnostics reveal that all our models provide a good fit to the data because the residuals are well behaved. A comparative analysis of the trigonometric quantile regression models with the unit generalized half-normal quantile regression model shows that the trigonometric models are the best. However, the sine unit generalized half-normal (SUGHN) quantile regression model is the best overall. It is observed that labor market insecurity and the homicide rate have significant negative effects on the educational attainment values of the OECD countries.

Fast and reliable prediction of scramjet flowfields via Gaussian process latent variable model and deep learning

Fast and accurate prediction of high-speed flowfields is of particular interest to researchers in fluid science and engineering to enable efficient design exploration and knowledge discovery. The reliability of prediction is another important metric for the performance of prediction models. While predictive modeling approaches with and without reduced-order modeling (ROM) via machine learning techniques have been proposed, they are inherently subject to loss of information for ROM-based approaches and substantial computational costs in modeling for non-ROM-based approaches. This paper proposes an accurate ROM-based predictive framework with minimum information loss enabled by incorporating Gaussian process latent variable modeling (GPLVM) and deep learning. The stochastic nature of GPLVM allows for uncertainty quantification that indicates the degree of prediction error or reliability of prediction without requiring validation data. The applicability for supersonic/hypersonic viscous flowfields has been examined for two cases including axisymmetric intakes and two-dimensional fuel injection in scramjet engines by comparison with other predictive models. Comparable or superior prediction accuracy over the other models has been achieved by the proposed approaches, demonstrating its high potential to serve as a new competent, data-driven technique for fast, accurate, and reliable prediction of scramjet flowfields.

Optimal redesign of a salinity monitoring network in Barataria Basin, Louisiana, USA using discrete entropy theory

We use discrete entropy theory and the MaxT model to optimize a continuous salinity monitoring network in a coastal Louisiana estuary (71 stations, 5 station types). Four station types represent marsh zones based on plant tolerance to salinity: fresh, intermediate, brackish, and saline. The fifth station type include stations in open water spanning the entire estuarine salinity gradient. A dry year (2012) with reduced freshwater inputs, and a wet year (2019) were chosen to test the robustness of the model. Our analysis showed that thirty-one stations formed the core network for both water years, with an additional 12 (2012) and 14 (2019) stations, unique for each year, needed to capture the optimal information for salinity in all five habitats. Fourteen stations could be eliminated with minimal loss of information regardless of whether the estuary was under drought or excess freshwater conditions. The Nash-Sutcliffe coefficient (ENs), coefficient of determination (R2) and the relative error (RE) showed good to excellent agreement between salinity measured by the original network and the modeled salinity of the modified network. Discrete entropy theory and MaxT proved effective in identifying redundancies in information sufficient to allow a 20% reduction of stations. Considering the highly variable nature of salinity within an estuary over short- and long-term temporal scales within a particular geographic location and across the entire estuary, this result was unexpected. Further reductions in stations (and corresponding savings in resources and costs) will require approaches not focused on replicating the exact information content from station to station but rather on measuring those salinity patterns and thresholds that shape biotic resource response in an estuary and possibly a reformulation of monitoring objectives. Monitoring objectives should include sampling for parameters (e.g., suspended sediments and nutrients) that don't require continuous data and are less costly to sustain yet provide equally critical information for understanding coastal wetland habitat sustainability and restoration.

Multi-site diagnostic classification of Autism spectrum disorder using adversarial deep learning on resting-state fMRI

Existing imaging studies in Autism spectrum disorder (ASD) mainly focused on single-site resting-state functional magnetic resonance imaging (rs-fMRI) data, which may contain limited samples and suffer from geographic bias. The lack of effective detection in information mining of rs-fMRI is a main reason for affecting the recognition rate of multi-site diagnostic. This study aims to propose a two-stage adversarial learning model with sliding window that integrates information from multi-site rs-fMRI data at the cost of minimal information loss. First, single rs-fMRI data is sampled with a sliding window to preserve both the spatial and temporal information of original data. And then, site-shared features of these samples are extracted through an adversarial learning model. Finally, the model is fine-tuned to learn discriminative disease-related features. Experimental results show that through adversarial learning, the heterogeneity problem among multi-site data is solved. Furthermore, the spatial–temporal information on rs-fMRI is effectively extracted and yields better classification performance (0.80 accuracy, 0.81 sensitivity, 0.80 specificity) than the state-of-the-art methods. Our results demonstrate the feasibility of the proposed method in the ASD classification task and the importance of fully exploiting the site-shared and spatial–temporal information in rs-fMRI data for multi-site ASD study.

A secure pixel level self-recovery scheme for digital images

This paper presents a secure image authentication scheme for tamper localization and recovery at pixel level. The proposed scheme encrypts the watermark comprising tamper localization code and self-recovery code using chaotic sequence to ensure security. This scheme uses pixel to block conversion technique for ensuring lossless recovery of the original image from an untampered watermarked image. For enhancing the localization accuracy, a multilevel tamper localization strategy is used. The experimental results show that the proposed scheme generates watermarked images with minimal information loss and can withstand copy-move, image splicing, content removal, vector quantization, collage and content only attacks. This scheme has better security, better tamper localization accuracy and better recovered image quality under extensive tampering and takes less computation time in comparison to the state-of-the-art schemes.

Modular Multi-Source Prediction of Drug Side-Effects With DruGNN.

Drug Side-Effects (DSEs) have a high impact on public health, care system costs, and drug discovery processes. Predicting the probability of side-effects, before their occurrence, is fundamental to reduce this impact, in particular on drug discovery. Candidate molecules could be screened before undergoing clinical trials, reducing the costs in time, money, and health of the participants. Drug side-effects are triggered by complex biological processes involving many different entities, from drug structures to protein-protein interactions. To predict their occurrence, it is necessary to integrate data from heterogeneous sources. In this work, such heterogeneous data is integrated into a graph dataset, expressively representing the relational information between different entities, such as drug molecules and genes. The relational nature of the dataset represents an important novelty for drug side-effect predictors. Graph Neural Networks (GNNs) are exploited to predict DSEs on our dataset with very promising results. GNNs are deep learning models that can process graph-structured data, with minimal information loss, and have been applied on a wide variety of biological tasks. Our experimental results confirm the advantage of using relationships between data entities, suggesting interesting future developments in this scope. The experimentation also shows the importance of specific subsets of data in determining associations between drugs and side-effects.

Efficient Congestion Control With Information Loss Minimization for Vehicular Ad Hoc Networks

The reduction of beacon message rates is of great significance to network congestion control in Vehicular Ad Hoc Networks (VANETs). However, this may cause large traffic information loss and serious traffic accidents. In this paper, we propose an efficient congestion control protocol, which not only alleviates channel congestion but also minimizes information loss by adaptively adjusting the beacon rates. Unlike conventional methods, this protocol investigates beacon transmission from a perspective of information loss, and adopts different beacon rates for different vehicles. First, we design an information loss factor to quantify the information deviation between the original information generated by a vehicle and the actual information received by its neighbor vehicles. Then, based on the information loss factor, we transform the beacon rate adjustment problem into an information loss minimization problem, which is addressed by a greedy beacon-rate adjustment algorithm. In addition, based on the derived different beacon rates, we design a periodicity-based time-slot selection scheme to allocate idle time slots to different vehicles, in order to alleviate transmission collisions. Extensive simulation results show the advantages of our solution in terms of multiple metrics.

A comparative analysis of the principal component method and parallel analysis in working with official statistical data

The dynamic development of the digitized society generates large-scale information data flows. Therefore, data need to be compressed in a way allowing its content to remain complete and informative. In order for the above to be achieved, it is advisable to use the principal component method whose main task is to reduce the dimension of multidimensional space with a minimal loss of information. The article describes the basic conceptual approaches to the definition of principle components. Moreover, the methodological principles of selecting the main components are presented. Among the many ways to select principle components, the easiest way is selecting the first k-number of components with the largest eigenvalues or to determine the percentage of the total variance explained by each component. Many statistical data packages often use the Kaiser method for this purpose. However, this method fails to take into account the fact that when dealing with random data (noise), it is possible to identify components with eigenvalues greater than one, or in other words, to select redundant components. We conclude that when selecting the main components, the classical mechanisms should be used with caution. The Parallel analysis method uses multiple data simulations to overcome the problem of random errors. This method assumes that the components of real data must have greater eigenvalues than the parallel components derived from simulated data which have the same sample size and design, variance and number of variables. A comparative analysis of the eigenvalues was performed by means of two methods: the Kaiser criterion and the parallel Horn analysis on the example of several data sets. The study shows that the method of parallel analysis produces more valid results with actual data sets. We believe that the main advantage of Parallel analysis is its ability to model the process of selecting the required number of main components by determining the point at which they cannot be distinguished from those generated by simulated noise.

Global optimal tracking control for multi-input-multi-output systems with adjoining cell mapping and a subdivision strategy

In this paper, global optimal tracking control for multi-input-multi-output systems using adjoining cell mapping technology is studied. An adaptive criterion is first proposed to improve the existing adjoining cell mapping method for optimal control. The criterion can judge the availability of mappings from different pre-image cells to their adjoining image cells adaptively, and help to build mapping database with the minimal loss of useful mapping information. An accessible and efficient backward search strategy to obtain global optimal control solution is also advanced. Then, a subdivision strategy for adjoining cell mapping to solve optimal tracking control problems for multi-input-multi-output systems is introduced. Finally, three examples illustrate the performance of the adaptive criterion and subdivision strategy. The adjoining cell mapping method for optimal control is greatly improved and the proposed subdivision strategy offers an efficient numerical method with general applicability to solve global optimal tracking control problems for nonlinear multi-input-multi-output systems.

Time-Ordered Recent Event (TORE) Volumes for Event Cameras.

Event cameras are an exciting, new sensor modality enabling high-speed imaging with extremely low-latency and wide dynamic range. Unfortunately, most machine learning architectures are not designed to directly handle sparse data, like that generated from event cameras. Many state-of-the-art algorithms for event cameras rely on interpolated event representations-obscuring crucial timing information, increasing the data volume, and limiting overall network performance. This paper details an event representation called Time-Ordered Recent Event (TORE) volumes. TORE volumes are designed to compactly store raw spike timing information with minimal information loss. This bio-inspired design is memory efficient, computationally fast, avoids time-blocking (i.e., fixed and predefined frame rates), and contains "local memory" from past data. The design is evaluated on a wide range of challenging tasks (e.g., event denoising, image reconstruction, classification, and human pose estimation) and is shown to dramatically improve state-of-the-art performance. TORE volumes are an easy-to-implement replacement for any algorithm currently utilizing event representations.

A Hybridized Artificial Neural Network for Automated Software Test Oracle

Software testing is the methodology of analyzing the nature of software to test if it works as anticipated so as to boost its reliability and quality. These two characteristics are very critical in the software applications of present times. When testers want to perform scenario evaluations, test oracles are generally employed in the third phase. Upon test case execution and test outcome generation, it is essential to validate the results so as to establish the software behavior’s correctness. By choosing a feasible technique for the test case optimization and prioritization as along with an appropriate assessment of the application, leads to a reduction in the fault detection work with minimal loss of information and would also greatly reduce the cost for clearing up. A hybrid Particle Swarm Optimization (PSO) with Stochastic Diffusion Search (PSO-SDS) based Neural Network, and a hybrid Harmony Search with Stochastic Diffusion Search (HS-SDS) based Neural Network has been proposed in this work. Further to evaluate the performance, it is compared with PSO- SDS based artificial Neural Network (PSO-SDS ANN) and Artificial Neural Network (ANN). The Misclassification of correction output (MCO) of HS-SDS Neural Network is 6.37 for 5 iterations and is well suited for automated testing.

DuctiLoc: Energy-Efficient Location Sampling With Configurable Accuracy

Mobile device tracking technologies based on various positioning systems have made location data collection ubiquitous. The frequency at which location samples are recorded varies across applications, yet it is usually pre-defined and fixed, resulting in redundant information, and draining the battery of mobile devices. In this paper, we first answer the question “ <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">at what frequency should individual human movements be sampled so that they can be reconstructed with minimum loss of information?</i> ”. Our analysis unveils a novel linear scaling law of the localization error with respect to the sampling interval. We then present DuctiLoc, a location sampling mechanism that utilises the law above to profile users and adapt the position tracking frequency to their mobility. DuctiLoc is energy efficient, as it does not rely on power-hungry sensors or expensive computations; moreover, it provides a handy knob to control energy usage, by configuring the target positioning accuracy. Controlling the trade-off between accuracy and sampling rate of human movement is useful in a number of contexts, including mobile computing and cellular networks. Real-world experiments with an Android implementation show that DuctiLoc can effectively adjust the sampling frequency to individual mobility habits and target accuracy level, reducing the energy consumption by 60% to 98% with respect to a baseline periodic sampling.