Synthetic Data Generation

Cutting-edge analytical instrumentation is increasingly being developed and applied to the analysis of fossils. X-ray fluorescence (XRF) imaging spectroscopy is a powerful tool to resolve the elemental chemistry of fossil specimens. Most of the XRF application to study fossils is carried out at dedicated synchrotron radiation XRF beamlines. Recent studies used laboratory scanners, i.e. stationary instruments with a measurement chamber or mobile ones to tackle paleontological questions. The application of these new XRF systems on fossils is still relatively limited and clear protocols for the acquisition and processing of the XRF data are currently lacking. Here, we present the use of mobile-macroscale scanning XRF (mobile-MA-XRF) imaging for the in situ analyses of the elemental chemistry of fossil vertebrates from the Messel biota (∼48 Ma, Eocene), including amphibians, reptiles, mammals and birds. We investigate the usefulness of mobile-MA-XRF to detect tissue-specific elemental signatures in fossils preserved in solid resin, liquid glycerin and water. We found remarkable tissue-specific chemical signatures preserved in almost all specimens analyzed. Hair and feathers are associated with S and Ti, abdominal tissues with Cu and Zn and stomach contents, e.g., seeds, are associated with Ni, Cu, and Zn. We provide a detailed protocol for acquisition and processing of MA-XRF data and a critical discussion of the application of this approach to paleontological research. Our work sets the foundation for applying MA-XRF to the analyses of those fossils that cannot be measured at synchrotron facilities and/or with stationary laboratory scanners due to their dimensions, weights and conservation mode.

A technique for estimating eddy diffusivities in a turbulent atmospheric layer is presented; the scheme adopted is based on an inverse-problem methodology. The inverse problem is formulated as a nonlinear constrained optimization problem, where the objective function is defined through the square differences between experimental and model data. The direct mathematical model is given by the advection-diffusion equation, which is solved by second-order finite-difference method. In the presence of noise it is necessary to use some regularization term; the Tikhonov function and an entropic regularization of zeroth, first and second orders are used in this paper. In addition, two inversion strategies are used: alternate and simultaneous eddy diffusivities estimation. Numerical experiments show a good performance of the proposed methodology.

In the present day digital world, it is imperative that all organizations and enterprises facilitate efficient processing of queries on XML data. XML queries typically specify patterns of selection predicates on multiple elements that have specified tree structured relationships. The primitive tree-structured relationships are parent-child and ancestordescendant. Finding all occurrences of these relationships in an XML database is a core operation for XML query processing. In this paper the pattern matching algorithms TwigStack and TwigStackList are discussed. The behavior of TwigStack is analyzed, and a comparison of these two algorithms is attempted. The TwigStack algorithm the initial holistic algorithm, has features of performing simultaneous scan over streams of XML nodes to match their structural relationships holistically, reducing a number of unnecessary intermediate results, and skipping XML nodes that will not contribute to final answers. The family of holistic pattern mat...

Algorithms for the inference of association with sequential information have been proposed and used but are ineffective, in some cases, because too many candidate rules are extracted. Filtering the relevant ones is usually difficult and inefficient. In this work, we present an algorithm for the inference of temporal association rules that uses context-free grammars to restrict the search process, in order to filter, in an efficient and effective way, the associations discovered by the algorithm. Moreover, we present experimental results that empirically evaluate its performance using synthetic data.

Completion of this thesis would not have been possible without the support and contribution of many people. It is a great honor for me to thank some of those many, to whom I owe my deepest gratitude. I would like to express my deepest appreciation to my supervisor, Dr. Sanjay Srinivasan for his excellent guidance, caring, patience, and providing me with an excellent atmosphere for doing research. He continually and convincingly conveyed a spirit of adventure in regard to research, and courage and versatility to battle challenging problems that I face in life. His scientific intuition, genius, invaluable insights, and continuous involvement were indispensable in developing my academic merits. Additional gratitude goes to Dr. Mrinal Sen for agreeing to be my secondary reader on such short notice. It gives me great pleasure in acknowledging the support of Dr. Kamy Sepehrnoori for helping me get through difficult times, and for all the emotional and technical guidance he provided during my life in Austin. I would also like to acknowledge the staff of the Department of Petroleum and

The Rover Environmental Monitoring Station (REMS) on the Mars Science Laboratory (MSL) offers the opportunity to explore the near surface atmospheric boundary layer (ABL) over an extended region of the Martian surface. The atmospheric boundary ...

We exhibit an option and more adaptable approach that amplifies client utility by fulfilling all clients. It does this while minimizing the utilization of framework assets. We examine the profits of this last approach and create a versatile checking result Satisfy User Profiles (Sups). Through formal dissection, we distinguish sufficient optimality conditions for SUP. Utilizing genuine (RSS channels) and manufactured follows, we exactly break down the conduct of SUP under shifting conditions. Our examinations demonstrate that we can attain a high level of fulfillment of client utility when the estimations of SUP nearly gauge the genuine occasion stream, and can possibly spare a lot of framework assets. We further demonstrate that SUP can misuse criticism to enhance client utility with just a moderate build in asset use.

Low-dimensional (2-or 3-dimensional) visual representations of large, highdimensional datasets with complicated cluster structures play a fundamental role in the discovery and identification of such structures. Visualization exploits the unmatched pattern recognition capability of humans for accurate and detailed cluster extraction, which is not possible with current automated methods because the latter still lack the power of the exceptional human reasoning. For explanatory and interactive visualization, a powerful tool is the use of self-organizing maps (SOMs). In general, by producing a spatially ordered set of quantization prototypes of large, higher-dimensional data, SOMs enable the visualization of various similarity information (such as prototype distances, distribution, topology) on a rigid lattice, without reducing the feature dimensionality. Information discovery further depends on the expressive power of the similarity measure and its visual representation. In this study, we compare the capabilities of our recent SOM visualization scheme, CONNvis, with prominent dimensionality reduction methods and show its superiority for visual assessment of intricate cluster structures.

An important piece of the ACARE (Advisory Council for Aeronautics in Europe) plan has been put in place early in 2005: the FLYSAFE Project (http://www.eu-flysafe.org/). FLYSAFE aims at defining and testing new tools and systems contributing to the safety of flights for all ...

Cubesats platforms expansion increases the need to simplify payloads and to optimize downlink data capabilities. A promising solution is to enhance on-board software, in order to take early decisions automatically. However, the most efficient methods for data analysis are generally large deep neural networks (DNN) oversized to be loaded and processed on limited hardware capacities of cubesats. To use them, we must reduce the size of DNN while accommodating efficiency in terms of both accuracy and inference cost. In this paper, we propose a distillation method which reduces image segmentation deep neural network's size to fit into on board processors. This method is presented through a ship detection example comparing accuracy and inference costs for several networks.

In this paper, a new iterative image processing algorithm is introduced and denoted as "iterative cellular image processing algorithm" (ICIPA). The new unsupervised iterative algorithm uses the advantage of stochastic properties and neighborhood relations between the cells of the input image. In ICIPA scheme; first regarding to the stochastic properties of the data, all possible quantization levels are determined and then 2D input image is processed using a function, based on averaging and neighborhood relationship, and after that a parameter C is assigned to each cell. Then Gaussian probability values are mapped to each cell regarding to all possible quantization levels and the attended value C. A maximum selector defines the highest probability value for each cell. In the case of complex data, first iteration output is fed into input till a sufficient output is found. We have applied ICIPA algorithm to various synthetic examples and then a real data, the ruins of Hittite Empire. Satisfactory results are obtained. We have observed that de-noising property of our scheme is the best in the literature. It is interesting that the corrupted data with Additive White Gaussian Noise (AWGN) up to 97% ratio, can be de-noised by using our proposed ICIPA algorithm.

We show that the problem of constructing a perfect matching in a graph is in the complexity class Random NC; i.e., the problem is solvable in polylog time by a randomized parallel algorithm using a polynomial -bounded number of processors. We also show that several related problems lie in Random NC.

Cet article aborde le problème du codage de séquences vidéo multi-vues et présente une nouvelle méthode d'estimation des champs de disparité et du mouvement impliqués dans la séquence. Afin de réduire la complexité et le coût de calcul et améliorer les performances d'estimation de ces champs, une nouvelle technique d'estimation conjointe est adoptée et appréhendée dans le cadre de méthodes variationnelles, en mettant en place un algorithme itératif, utilisant des outils récents d'analyse convexe. Les expérimentations sur des séquences réelles montrent les performances de cette technique aussi bien en termes de reconstruction qu'en termes de cohérence des champs de déplacement estimés.

A novel variable influence on projection approach for O2PLS models, named VIP O2PLS , is presented in this paper. VIP O2PLS is a model-based method for judging the importance of variables. Its cornerstone is the 2-way formalism of the O2PLS models; i.e. the use of both predictive and orthogonal normalized loadings of the two modelled data matrices, and also a new weighting system based on the sum of squares of both data blocks (X, Y). The VIP O2PLS algorithm has been tested in one synthetic data set and two real cases, and the outcomes have been compared to the PLS-VIP, VIP OPLS , and i-PLS methods. The purpose is to achieve a sharper and enhanced model interpretation of O2PLS models by using the new VIP O2PLS method for assessing the importance of both X-and Y-variables.

Euler deconvolution is a commonly employed magnetic interpretation method because it requires only a little a priori knowledge about the magnetic source geometry, and, more importantly, because it requires no information about the magnetization vector. As a result, it may be successfully applied in areas where the geology is poorly known. However, it requires a priori knowledge about the nature of an equivalent source producing a magnetic anomaly with the same falloff rate of the observed anomaly. This is a crucial limitation of the method, requiring that a parameter known as structural index Ž . h be determined. The customary application of the Euler method and the process of estimating h are benefited by having a large number of data and solutions preventing its application to ground magnetic surveys which may consist of a limited number of observations. We show that if the structural index is estimated by a new criterion, Euler deconvolution becomes a feasible technique to interpret anomalies defined by just a few observations. This new criterion is based on the correlation between the total-field anomaly h o and the estimates of the base level b. These estimates are obtained for each position of the moving data window along the observed profile and for several tentative values for the structural index. However, differently from the customary method, instead of estimating the structural index as the tentative value producing the smallest solution dispersion, the best estimate of h is taken as the tentative value leading to the smallest correlation between h o and the estimates of b. This criterion is deduced from the Euler's equation, so it does not depend on the inclination and declination of the geomagnetic field. The good results obtained with this new criterion in determining the correct structural index is illustrated in tests using synthetic data from different latitudes and the feasibility in using this criterion in applying Euler deconvolution to ground surveys is illustrated with a real magnetic anomaly defined by just 12 observations and produced by a basicrultrabasic body at the emerald deposit of Socoto, Bahia, Brazil. The results of Euler deconvolution, ćombined with the geological information that the basicrultrabasic body outcrops, show that the intrusive body may be approximated by an outcropping horizontal cylinder with a diameter of 68 m and center at a depth of 34 m, which is consistent with the geologic knowledge of the deposit.

We have developed a new approach for estimating the location and geometry of several density anomalies that give rise to a complex, interfering gravity field. The user interactively defines the assumed outline of the true gravity sources in terms of points and line segments, and the method estimates sources closest to the specified outline to achieve a match between the predicted and observed gravity fields. Each gravity source is assumed to be a homogeneous body with a known density contrast; different density contrasts may be assigned to each source. Tests with synthetic data show that the method can be of use in estimating (1) multiple laterally adjacent and closely situated gravity sources, (2) single gravity sources consisting of several homogeneous compartments with different density contrasts, and (3) two gravity sources with different density contrasts of the same sign, one totally enclosed by the other. The method is also applied to three different sets of field data where ...

Prototype databases are needed in any information system development process to support data-intensive applications development. It is common practice to populate these databases using synthetic data. This data usually bears little relation to the application's domain and considers only a very reduced subset of the integrity constraints the database will hold during operation. This paper claims that in situations where operational data is available, as is the case in information system evolution and migration, a sample of this data must be used to create a prototype database. The paper outlines a method for consistently sampling a database. This method uses a new concept, the Insertions Chain Graph, to assist in selecting instances so that the resulting Sample Database reaches a consistent state, a signi cant task of database sampling.

Automated feature engineering (AutoFE) has become a cornerstone of efficient machine learning (ML), yet its potential to perpetuate or amplify bias remains underexplored. This paper proposes a fairness-aware framework for feature transformation, addressing how AutoFE tools-while optimizing for model performance-may inadvertently encode discriminatory patterns into derived features. Drawing on Ferrario et al. (2022)'s work on bias propagation in ML pipelines and Kamiran & Calders (2019)'s foundational methods for discrimination-aware data mining, we first demonstrate that common AutoFE techniques (e.g., feature synthesis, aggregation) can systematically marginalize underrepresented groups by reinforcing spurious correlations. We then introduce FairFeature, a novel framework that integrates bias metrics (e.g., demographic parity, equalized odds) directly into the feature generation process. Unlike post-hoc fairness adjustments (e.g., adversarial debiasing), FairFeature proactively constrains feature transformations using fairness-aware optimization, ensuring that engineered features meet both predictive utility and equity criteria. Empirical evaluations on real-world datasets (e.g., UCI Adult, COMPAS) reveal that AutoFE without fairness constraints increases disparity by up to 22% in model outcomes, while FairFeature reduces bias by 35-60% with <5% accuracy trade-offs. Our work bridges critical gaps between data engineering and algorithmic fairness, offering practitioners a scalable tool to mitigate 62 hidden biases at the feature level. We further release an open-source library implementing FairFeature to foster adoption.

Automated feature engineering (AutoFE) has become a cornerstone of efficient machine learning (ML), yet its potential to perpetuate or amplify bias remains underexplored. This paper proposes a fairness-aware framework for feature transformation, addressing how AutoFE tools-while optimizing for model performance-may inadvertently encode discriminatory patterns into derived features. Drawing on Ferrario et al. (2022)'s work on bias propagation in ML pipelines and Kamiran & Calders (2019)'s foundational methods for discrimination-aware data mining, we first demonstrate that common AutoFE techniques (e.g., feature synthesis, aggregation) can systematically marginalize underrepresented groups by reinforcing spurious correlations. We then introduce FairFeature, a novel framework that integrates bias metrics (e.g., demographic parity, equalized odds) directly into the feature generation process. Unlike post-hoc fairness adjustments (e.g., adversarial debiasing), FairFeature proactively constrains feature transformations using fairness-aware optimization, ensuring that engineered features meet both predictive utility and equity criteria. Empirical evaluations on real-world datasets (e.g., UCI Adult, COMPAS) reveal that AutoFE without fairness constraints increases disparity by up to 22% in model outcomes, while FairFeature reduces bias by 35-60% with <5% accuracy trade-offs. Our work bridges critical gaps between data engineering and algorithmic fairness, offering practitioners a scalable tool to mitigate 62 hidden biases at the feature level. We further release an open-source library implementing FairFeature to foster adoption.

We study recovery conditions of weighted 1 minimization for signal reconstruction from compressed sensing measurements when partial support information is available. We show that if at least 50% of the (partial) support information is accurate, then weighted 1 minimization is stable and robust under weaker sufficient conditions than the analogous conditions for standard 1 minimization. Moreover, weighted 1 minimization provides better upper bounds on the reconstruction error in terms of the measurement noise and the compressibility of the signal to be recovered. We illustrate our results with extensive numerical experiments on synthetic data and real audio and video signals.

The Rover Environmental Monitoring Station (REMS) on the Mars Science Laboratory (MSL) offers the opportunity to explore the near surface atmospheric boundary layer (ABL) over an extended region of the Martian surface. The atmospheric boundary ...

Towed streamer electromagnetic (TSEM) survey is an efficient data acquisition technique capable of collecting a large volume of electromagnetic (EM) data over extensive areas rapidly and economically. The TSEM survey is capable of detecting and characterizing marine hydrocarbon (HC) reservoirs. However, interpretation of the TSEM data is still a very challenging problem. We propose solving this problem by migrating the optimal synthetic aperture (OSA) data for the TSEM survey. We first represent the OSA data as a solution of Lippmann-Schwinger equation and then demonstrate that the migration of OSA data is just the inner product of the backwardpropagated and forward-propagated EM fields. The migration problem is solved iteratively within the general framework of the reweighted, regularized, conjugate gradient (RRCG) method. The proposed method was tested with two synthetic models. We also applied this method to the TSEM data set collected in the Barents Sea and revealed a resistive layer at a depth of about 500 m.

This paper describes an application, called Medici, designed to produce synthetic data for social network graphs, which can be used for analysis, hypothesis testing and application development by researchers and practitioners in the field. It builds on previous work by providing an integrated system, and a user friendly screen interface. It can be run with default values to produce graph data and statistics, which can then be used for further processing. The system is made publicly available in a Github Java project. The annex provides a user manual with a screen by screen guide.

Two of the difficulties for data analysts of online social networks are (i) the public availability of data and (ii) respecting the privacy of the users. One possible solution to both of these problems is to use synthetically generated data. However, this presents a series of challenges related to generating a realistic dataset in terms of topologies, attribute values, communities, data distributions, correlations and so on. In the following work we present and validate an approach for populating a graph topology with synthetic data which approximates an online social network. The empirical tests confirm that our approach generates both a diverse and a correlated dataset, with a realistic modeling of noise and interactions between communities. The data generator is also highly configurable, with a sophisticated control parameter set for different "dispersion" levels.

In probabilistic mobile robotics, the development of measurement models plays a crucial role as it directly influences the efficiency and the robustness of the robot's performance in a great variety of tasks including localization, tracking, and map building. In this paper, we present a novel probabilistic measurement model for range finders, called Gaussian beam processes, which treats the measurement modeling task as a nonparametric Bayesian regression problem and solves it using Gaussian processes. The major benefit of our approach is its ability to generalize over entire range scans directly. This way, we can learn the distributions of range measurements for whole regions of the robot's configuration space from only few recorded or simulated range scans. Especially in approximative approaches to state estimation like particle filtering or histogram filtering, this leads to a better approximation of the true likelihood function. Experiments on real world and synthetic data show that Gaussian beam processes combine the advantages of two popular measurement models.

This paper describes the development and application of a system for quantitative near-bottom seismic profiling at 4 kHz. Developed as part of the Deep-Tow Instrumentation System of the Marine Physical Laboratory, this system represents a capability which is relatively unique in terms of lateral resolution. This capability was developed in order to observe small-scale horizontal variations in acoustic reflectivity of the sea floor as well as similar properties for shallow, layered structures in bottom sediments. The system provides real-time computer processing and quantitative displays of processed data in several forms. These displays include a high-resolution grey scale display of equivalent plane wave pressure (compensating for Fish altitude), a three-dimensional contourogram display of equivalent intensity, and a plot of integrated equivalent energy for travel time sections (typically corresponding to 0–5 and 5–55 m into the bottom). These quantitative displays provide a real-t...

This data paper presents lightcurves of 101 near Earth asteroids (NEAs) observed mostly between 2014 and 2017 as part of the EURONEAR photometric survey using 11 telescopes with diameters between 0.4 and 4.2 m located in Spain, Chile, Slovakia and Romania. Most targets had no published data at the time of observing, but some objects were observed in the same period mainly by B. Warner, allowing us to confirm or improve the existing results. To plan the runs and select the targets, we developed the public Long Planning tool in PHP. For preliminary data reduction and rapid follow-up planning we developed the LiDAS pipeline in Python and IRAF. For final data reduction, flux calibration, night linkage and Fourier fitting, we used mainly MPO Canopus.

The potentials around a finite cylindrical electrode can be obtained by dividing the electrodes into rings of equal thickness and substituting an infinitely thin current ring for each of the slices. The field of an infinitely thin ring electrode mounted on an insulating cylindrical probe of the same diameter can be found by combining the properties of the delta function with a solution of Laplace' s equation in cylindrical coordinates. Combination of solutions for the infinitely thin rings under the condition that the potential of the electrode surface be constant leads to a system of simultaneous linear equations. By increasing the number of slices, the potential around the finite electrode can be found arbitrarily close. The problem of a cylindrical electrode on a sonde located coaxially in a conducting hole, drilled through a medium of different conductivity, is treated by the same method. This arrangement is of interest in electrical logging of drill holes. Numerical examples have been calculated on an IBM 650 magnetic drum computer. The potential along the surface of the insulating probe, at distances larger than twice the electrode length, can be approximated with good accuracy by assuming that all of the current is emitted from an infinitely thin ring located in the median plane of the electrode.

The liberation distribution of ore samples is of considerable interest for process optimisation in the minerals industry. A scanning electron microscope-based automatic mineral analyser such as the LEO QEMSCAN system developed by CSIRO Minerals is a powerful tool for the estimation of linear or areal grade distributions of a population of ore particles based upon polished single particle sections. Stereological correction of a single section mineralogical grade distribution is recognised as an ill-posed inverse problem. The transformation kernel method with constrained entropy regularisation (King and Schneider, 1998) is adopted for the correction of stereological error in binary systems. An enhanced transformation kernel correction scheme is developed with an additional equality constraint for average grade as determined by section and volumetric sampling, in accordance with Delesse&#39;s fundamental stereological theorem. The usefulness of both correction methods is limited by the...

Cortical events of correlated neuronal firing are thought to underlie sensorimotor and associative functions. It is believed that events are attractors of cortical dynamics, pulled by strong mutual connections between recurrently active -core -neurons. Here, we exhaustively measured core structural connectivity, using a unique dataset with co-registered two-photon calcium imaging and electron microscopy of the same cortical tissue. We report that cores make fewer weaker mutual connections compared to other neurons. Instead, cores have connections arranged to funnel incoming activity. With a simple model, we advance that a common distance-dependent connectivity rule yields cores with funneling connections supporting population events. Our integrated analysis shows that cortical attractor-like dynamics cannot abstract from the identity of, and connections between, the specific neurons implementing them.

This paper examines the impact of a mesoscale analysis (2.5 km grid distance) on the simulation of the meso-gamma scale aspects of föhn in the Rhine Valley. The föhn event, documented during IOP 15 (5 November 1999) of the Mesoscale Alpine Programme, was standard in terms of intensity and was characterized by an important temporal variability. Many instruments operating in the Rhine Valley target area are used to validate the simulation, in particular the airborne nadir pointing lidar LEANDRE 2 (flown over the Lower Rhine Valley) as well as a windprofiler and a radio accoustic sounding system colocated in Rankweil, Austria. The large observational dataset acquired during the IOP allowed documentation of the entire föhn life cycle. For most of the IOP, a cold pool remained near the ground in the lower northern part of the valley. The non-hydrostatic model meso-NH, used in a grid-nesting configuration with two nested models and initialised with a mesoscale analyse, allowed to simulate realistically the location and depth of the cold pool. The relationship between the föhn intensity and the large scale environment is also examined. The flow regime is a flow around the Alps. The variability of this flow at the western tip of the Alps could explain some of the temporal changes observed at low level in the Rhine valley.

The description that one can have of the seismic source is the mani- festation of an imagined model, obviously outlined from Physic Theories and supported by mathematical methods. In that context, the modelling of earthquake rupture consists in finding values of the parameters of the selected physics-mathematical model, through which it becomes possible to reproduce numerically the records of earthquake effects on the Earths surface. We present and test a Linear Programming (LP) inversion in dual form, for reconstructing the kinematics of the rupture of large earthquakes through space-time seismic slip distribution on finite faults planes

We investigate the ability of a mesoscale model to reconstruct CO 2 fluxes at regional scale. Formally, we estimate the reduction of error for a CO 2 flux inversion at 8 km resolution in the South West of France, during four days of the CarboEurope Regional Experiment Strategy (CERES) in spring 2005. Measurements from two towers and two airplanes are available for this campaign. The lagrangian particle dispersion model LPDM was coupled to the nonhydrostatic model Meso-NH and integrated in a matrix inversion framework. Impacts of aircraft and tower measurements are quantified separately and together. We find that the configuration with both towers and aircraft is able to significantly reduce uncertainties on the 4-day averaged CO 2 fluxes over about half of the 300×300 km 2 domain. Most of this reduction comes from the tower measurements, even though the impact of aircraft measurements remains noticeable. Imperfect knowledge of boundary conditions does not significantly impact the error reduction for surface fluxes. We test alternative strategies to improve the impact of aircraft measurements and find that most information comes from measurements inside the boundary layer. We find that there would be a large improvement in error reduction if we could improve our ability to model nocturnal concentrations at tower sites.

We consider the problem of computing a minimum-weight polygonal path between two points in a weighted polygonal subdivision, subject to the constraint that the path have few segments (links). We give an algorithm that generates paths of weighted length at most (1+ ) times the weight of a minimum-cost k-link path, for any fixed > 0, while using at most 2k -1 links. This is an improvement over the previous (1 + )-approximation algorithm, which used at most 5k -2 links. Further, we have implemented our new algorithm and we have conducted a performance study of these algorithms (old and new) on a variety of real-world and synthetic data, comparing not only the efficiency but also the quality of paths generated using these algorithms. We also consider the implications of these results on the practical usage of these algorithms.

Mushroom is one of the fungi types' food that has the most potent nutrients on the plant. Mushrooms have major medical advantages such as killing cancer cells. This study aims to find the most appropriate technique for mushroom classification, and mushroom will be classified into two categories, poisonous and nonpoisonous. The proposed approach will implement a different techniques and algorithms like neural network (NN), Support Vector Machines (SVM), Decision Tree, and k Nearest Neighbors (KNN), on dataset of mushroom images, where the dataset contains images with background and without background. The experimental results shown that the best technique for classifying mushroom images is kNN with accuracy of 94% based on features extracted from images with real dimensions of mushroom types, and 87% based on features extracted from images only.

Synthetic datasets are beneficial for machine learning researchers due to the possibility of experimenting with new strategies and algorithms in the training and testing phases. These datasets can easily include more scenarios that might be costly to research with real data or can complement and, in some cases, replace real data measurements, depending on the quality of the synthetic data. They can also solve the unbalanced data problem, avoid overfitting, and can be used in training while testing can be done with real data. In this paper, we present, to the best of our knowledge, the first synthetic dataset for Unmanned Aerial Vehicle (UAV) attacks in 5G and beyond networks based on the following key observable network parameters that indicate power levels: the Received Signal Strength Indicator (RSSI) and the Signal to Interference-plus-Noise Ratio (SINR). The main objective of this data is to enable deep network development for UAV communication security. Especially, for algorithm development or the analysis of time-series data applied to UAV attack recognition. Our proposed dataset provides insights into network functionality when static or moving UAV attackers target authenticated UAVs in an urban environment. The dataset also considers the presence and absence of authenticated terrestrial users in the network, which may decrease the deep network's ability to identify attacks. Furthermore, the data provides deeper comprehension of the metrics available in the 5G physical and MAC layers for machine learning and statistics research. The dataset will available at link

Re ´sume ´: Dans ce rapport, nous conside ´rons une paire d'images ste ´re ´o dont seule la ge ´ome ´trie e ´pipolaire est connue, repre ´sente ´e par la .matrice fondamentale du syste `me de came ´ras. Nous montrons qu'il est possible de de ´terminer directement "dans les images", c'est-a `-dire sans effectuer de reconstruction tridimensionnelle, des proprie ´te ´s tridimensionnelles importantes et intuitives comme la position relative de points et de plans ou l'enveloppe convexe d'un ensemble de points. Nous introduisons la notion de visibilite ´, qui permet d'obtenir ces proprie ´te ´s. Nous pre ´sentons des re ´sultats obtenus sur des donne ´es re ´elles.

Data on businesses collected by statistical agencies are challenging to protect. Many businesses have unique characteristics, and distributions of employment, sales, and profits are highly skewed. Attackers wishing to conduct identification attacks often have access to much more information than for any individual. As a consequence, most disclosure avoidance mechanisms fail to strike an acceptable balance between usefulness and confidentiality protection. Detailed aggregate statistics by geography or detailed industry classes are rare, public-use microdata on businesses are virtually inexistant, and access to confidential microdata can be burdensome. Synthetic microdata have been proposed as a secure mechanism to publish microdata, as part of a broader discussion of how to provide broader access to such data sets to researchers. In this article, we document an experiment to create analytically valid synthetic data, using the exact same model and methods previously employed for the U...

We present a discriminative learning framework for Gaussian mixture models (GMMs) used for classification based on the extended Baum-Welch (EBW) algorithm . We suggest two criteria for discriminative optimization, namely the class conditional likelihood (CL) and the maximization of the margin (MM). In the experiments, we present results for synthetic data, broad phonetic classification, and a remote sensing application. The experiments show that CL-optimized GMMs (CL-GMMs) achieve a lower performance compared to MM-optimized GMMs (MM-GMMs), whereas both discriminative GMMs (DGMMs) perform significantly better than generatively learned GMMs. We also show that the generative discriminatively parameterized GMM classifiers still allow to marginalize over missing features, a case where generative classifiers have an advantage over purely discriminative classifiers such as support vector machines or neural networks.

Engineering Systems for Knowledge and Inference, an emerging focus area for Engineering as well as for the country at large, encompasses a wide variety of technologies. The goal is to generate new understanding or knowledge of situations, thereby allowing anticipation or prediction of possible outcomes. With this knowledge, a more comprehensive solution may be possible for problems as complex as the prediction of disease outbreaks or advance warning of terrorist threats. FY2006 LDRD research included projects to determine the location and contents of atmospheric releases based on sensors in the fi eld, and new research into decomposing extremely large semantic graphs. Tech Base efforts included a testbed to evaluate hierarchical clustering, and analysis tools and application of a previously developed image content engine to new inspection capabilities in optics. Energy Manipulation, a long time focus that is receiving increased emphasis due to newly emerging applications, encompasses the fundamental understanding and technology deployment for many modern pulsedpower applications. This area has broad applications for magnetic fl ux compression generators and components for modern accelerators. FY2006 LDRD research included research into the fundamental mechanisms responsible for fl ashover of insulators subject to microsecond pulses. Tech Base efforts involved capabilities to improve the switching performance of power MOSFETs, and solid-state replacements for ignitrons.

Using multiyear satellite rainfall estimates, the distributions of the area, and the total rain rate of rain clusters over the equatorial Indian, Pacific, and Atlantic Oceans was found to exhibit a power law f S s ð Þ e s Àζ S , in which S represents either the cluster area or the cluster total rain rate and f S (s) denotes the probability density function of finding an event of size s. The scaling exponents ζ S were estimated to be 1.66 ± 0.06 and 1.48 ± 0.13 for the cluster area and cluster total rain rate, respectively. The two exponents were further found to be related via the expected total rain rate given a cluster area. These results suggest that convection over the tropical oceans is organized into rain clusters with universal scaling properties. They are also related through a simple scaling relation consistent with classical self-organized critical phenomena. The results from this study suggest that mesoscale rain clusters tend to grow by increasing in size and intensity, while larger clusters tend to grow by self-organizing without intensification.

Starting from the definition of Azimuth Moveout (AM O) as the cascade of D M O and inverse D M O at different offsets and azimuths, we derive an amplitude-preserving function for the AM O operator. This amplitude function is based on the FK definition of D M O and the definition of its true inverse. Similar to Liner's formalism of a true inverse for Hale's D M O, we derive an asymptotically true inverse for Black/Zhang's D M O and Bleinstein's Born D M O. A numerical test is given that compares amplitude preservation using kinematically equivalent D M O operators cascaded with their true inverses. We define amplitude-preserved processing as the preservation of the offset-dependent reflectivity after AM O transformation, where the reflectivity is considered to be proportional to the peak amplitude of each event. We found that an AM O operator defined using Zhang's D M O cascaded with its true inverse best reconstructs data amplitudes after transformation to a new offset and azimuth. The new amplitude function represents an amplitudepreserving azimuth moveout.

In the context of deep geological disposal of high level radioactive wastes, the French National Radioactive Waste Management Agency (Andra) has conducted an extensive characterization of the Callovo-Oxfordian argillaceous rock and surrounding formations in the Eastern Paris Basin. For such characterization, a detailed 3D geological model is needed. This paper shows the procedure used for building the 3D model with depth by combining time-to-depth conversion of seismic horizons, consistent seismic velocity model and elastic impedance in time. It also shows how the 3D model is used for mechanical and hydrogeological studies. The 3D field data example illustrates the potential of the proposed depth conversion procedure for estimating a density model with depth. The geological model shows good agreement with logging data obtained at a reference well, a closest borehole located in the vicinity of the 3D seismic area. Modeling of mechanical parameters such as shear modulus, young modulus...

Acoustic sonar imaging systems are widely used for underwater surveillance in both civilian and military sectors. However, acquiring high-quality sonar datasets for training Artificial Intelligence (AI) models confronts challenges such as limited data availability, financial constraints, and data confidentiality. To overcome these challenges, we propose a novel benchmark dataset of Simulated Side-Scan Sonar images, which we term as 'S3Simulator dataset'. Our dataset creation utilizes advanced simulation techniques to accurately replicate underwater conditions and produce diverse synthetic sonar imaging. In particular, the cutting-edge AI segmentation tool i.e. Segment Anything Model (SAM) is leveraged for optimally isolating and segmenting the object images, such as ships and planes, from real scenes. Further, advanced Computer-Aided Design tools i.e. SelfCAD and simulation software such as Gazebo are employed to create the 3D model and to optimally visualize within realistic environments, respectively. Further, a range of computational imaging techniques are employed to improve the quality of the data, enabling the AI models for the analysis of the sonar images. Extensive analyses are carried out on S3simulator as well as real sonar datasets to validate the performance of AI models for underwater object classification. Our experimental results highlight that the S3Simulator dataset will be a promising benchmark dataset for research on underwater image analysis. https://github.com/bashakamal/S3Simulator.

In this paper, we aim to identify the minimal subset of discrete random variables that is relevant for probabilistic classification in data sets with many variables but few instances. A principled solution to this problem is to determine the Markov boundary of the class variable. Also, we present a novel scalable, data efficient and correct Markov boundary learning algorithm under the so-called faithfulness condition. We report extensive empiric experiments on synthetic and real data sets scaling up to 139,351 variables.