Search results for: Stack Generalization

Authors: M. Hussaina, K. Mahboobb, S. Z. Ilyasa, S. Shaheena

Abstract:

Plume dispersion modeling is a computational procedure to establish a relationship between emissions, meteorology, atmospheric concentrations, deposition and other factors. The emission characteristics (stack height, stack diameter, release velocity, heat contents, chemical and physical properties of the gases/particle released etc.), terrain (surface roughness, local topography, nearby buildings) and meteorology (wind speed, stability, mixing height, etc.) are required for the modeling of the plume dispersion and estimation of ground and air concentration. During the early phase of Fukushima accident, plume dispersion modeling and decisions were taken for the implementation of protective measures. A difference in estimated results and decisions made by different countries for taking protective actions created a concern in local and international community regarding the exact identification of the safe zone. The current study is focused to highlight the importance of accurate and exact weather data availability, scientific approach for decision making for taking urgent protective actions, compatible and harmonized approach for plume dispersion modeling during a nuclear emergency. As a case study, the influence of meteorological data on plume dispersion modeling and decision-making process has been performed.

Keywords: decision making process, radiation doses, nuclear emergency, meteorological implications

Procedia PDF Downloads 165

Authors: Flavien Marteau, Pedro Affonso Nóbrega, Pascal Biwole, Nicolas Autrusson, Iona De Bievre, Christian Beauger

Abstract:

Effective water management is essential for the optimal performance of fuel cells. For this reason, many vehicle systems use a membrane humidifier, a passive device that humidifies the air before the cathode inlet. Although they offer good performance, humidifiers are voluminous, costly, and fragile, hence the desire to find an alternative. Direct water injection could be an option, although this method lacks maturity. It consists of injecting liquid water as a spray in the dry heated air coming out from the compressor. This work focuses on the evaluation of direct water injection and its performance compared to the membrane humidifier selected as a reference. Two architectures were experimentally tested to humidify an industrial 2 kW short stack made up of 20 cells of 150 cm² each. For the reference architecture, the inlet air is humidified with a commercial membrane humidifier. For the direct water injection architecture, a pneumatic nozzle was selected to generate a fine spray in the air flow with a Sauter mean diameter of about 20 μm. Initial performance was compared over the entire range of current based on polarisation curves. Then, the influence of various parameters impacting water management was studied, such as the temperature, the gas stoichiometry, and the water injection flow rate. The experimental results obtained confirm the possibility of humidifying the fuel cell using direct water injection. This study, however shows the limits of this humidification method, the mean cell voltage being significantly lower in some operating conditions with direct water injection than with the membrane humidifier. The voltage drop reaches 30 mV per cell (4 %) at 1 A/cm² (1,8 bara, 80 °C) and increases in more demanding humidification conditions. It is noteworthy that the heat of compression available is not enough to evaporate all the injected liquid water in the case of DWI, resulting in a mix of liquid and vapour water entering the fuel cell, whereas only vapour is present with the humidifier. Variation of the injection flow rate shows that part of the injected water is useless for humidification and seems to cross channels without reaching the membrane. The stack was successfully humidified thanks to direct water injection. Nevertheless, our work shows that its implementation requires substantial adaptations and may reduce the fuel cell stack performance when compared to conventional membrane humidifiers, but opportunities for optimisation have been identified.

Keywords: cathode humidification, direct water injection, membrane humidifier, proton exchange membrane fuel cell

Procedia PDF Downloads 22

Authors: Adesoji Tunbosun Jaiyeola, Josiah Adeyemo

Abstract:

Sediment yield can be considered to be the total sediment load that leaves a drainage basin. The knowledge of the quantity of sediments present in a river at a particular time can lead to better flood capacity in reservoirs and consequently help to control over-bane flooding. Furthermore, as sediment accumulates in the reservoir, it gradually loses its ability to store water for the purposes for which it was built. The development of hydrological models to forecast the quantity of sediment present in a reservoir helps planners and managers of water resources systems, to understand the system better in terms of its problems and alternative ways to address them. The application of artificial intelligence models and technique to such real-life situations have proven to be an effective approach of solving complex problems. This paper makes an extensive review of literature relevant to the theories and applications of evolutionary algorithms, and most especially genetic programming. The successful applications of genetic programming as a soft computing technique were reviewed in sediment modelling and other branches of knowledge. Some fundamental issues such as benchmark, generalization ability, bloat and over-fitting and other open issues relating to the working principles of GP, which needs to be addressed by the GP community were also highlighted. This review aim to give GP theoreticians, researchers and the general community of GP enough research direction, valuable guide and also keep all stakeholders abreast of the issues which need attention during the next decade for the advancement of GP.

Keywords: benchmark, bloat, generalization, genetic programming, over-fitting, sediment yield

Procedia PDF Downloads 427

Authors: Yehjune Heo

Abstract:

Fingerprint Anti-Spoofing approaches have been actively developed and applied in real-world applications. One of the main problems for Fingerprint Anti-Spoofing is not robust to unseen samples, especially in real-world scenarios. A possible solution will be to generate artificial, but realistic fingerprint samples and use them for training in order to achieve good generalization. This paper contains experimental and comparative results with currently popular GAN based methods and uses realistic synthesis of fingerprints in training in order to increase the performance. Among various GAN models, the most popular StyleGAN is used for the experiments. The CNN models were first trained with the dataset that did not contain generated fake images and the accuracy along with the mean average error rate were recorded. Then, the fake generated images (fake images of live fingerprints and fake images of spoof fingerprints) were each combined with the original images (real images of live fingerprints and real images of spoof fingerprints), and various CNN models were trained. The best performances for each CNN model, trained with the dataset of generated fake images and each time the accuracy and the mean average error rate, were recorded. We observe that current GAN based approaches need significant improvements for the Anti-Spoofing performance, although the overall quality of the synthesized fingerprints seems to be reasonable. We include the analysis of this performance degradation, especially with a small number of samples. In addition, we suggest several approaches towards improved generalization with a small number of samples, by focusing on what GAN based approaches should learn and should not learn.

Keywords: anti-spoofing, CNN, fingerprint recognition, GAN

Procedia PDF Downloads 173

Authors: Preeti Sharma

Abstract:

This paper deals with the Stancu type generalization of Lupaş-Durrmeyer operators. We establish some direct results in the polynomial weighted space of continuous functions defined on the interval [0, 1]. Also, Voronovskaja type theorem is studied.

Keywords: Lupas-Durrmeyer operators, polya distribution, weighted approximation, rate of convergence, modulus of continuity

Procedia PDF Downloads 326

Authors: Varun Agarwal

Abstract:

Introduction: With the advent of personalized medicine, histopathological review of whole slide images (WSIs) for cancer diagnosis presents an exceedingly time-consuming, complex task. Specifically, detecting metastatic regions in WSIs of sentinel lymph node biopsies necessitates a full-scanned, holistic evaluation of the image. Thus, digital pathology, low-level image manipulation algorithms, and machine learning provide significant advancements in improving the efficiency and accuracy of WSI analysis. Using Camelyon16 data, this paper proposes a deep learning pipeline to automate and ameliorate breast cancer metastasis localization and WSI classification. Methodology: The model broadly follows five stages -region of interest detection, WSI partitioning into image tiles, convolutional neural network (CNN) image-segment classifications, probabilistic mapping of tumor localizations, and further processing for whole WSI classification. Transfer learning is applied to the task, with the implementation of Inception-ResNetV2 - an effective CNN classifier that uses residual connections to enhance feature representation, adding convolved outputs in the inception unit to the proceeding input data. Moreover, in order to augment the performance of the transfer learning CNN, a stack of convolutional denoising autoencoders (CDAE) is applied to produce embeddings that enrich image representation. Through a saliency-detection algorithm, visual training segments are generated, which are then processed through a denoising autoencoder -primarily consisting of convolutional, leaky rectified linear unit, and batch normalization layers- and subsequently a contrast-normalization function. A spatial pyramid pooling algorithm extracts the key features from the processed image, creating a viable feature map for the CNN that minimizes spatial resolution and noise. Results and Conclusion: The simplified and effective architecture of the fine-tuned transfer learning Inception-ResNetV2 network enhanced with the CDAE stack yields state of the art performance in WSI classification and tumor localization, achieving AUC scores of 0.947 and 0.753, respectively. The convolutional feature retention and compilation with the residual connections to inception units synergized with the input denoising algorithm enable the pipeline to serve as an effective, efficient tool in the histopathological review of WSIs.

Keywords: breast cancer, convolutional neural networks, metastasis mapping, whole slide images

Procedia PDF Downloads 113

Authors: Benedictus Asriparusa

Abstract:

In the area of the oil industry, there is accompanied by associated natural gas. The thing shows that a large amount of energy is being wasted mostly in the developing countries by contributing to the global warming process. This research represents an overview of methods in Minas area employed by these researchers in PT. Chevron Pacific Indonesia to determine ways of measuring and reducing gas flaring and its emission drastically. It provides an approximation includes analytical studies, numerical studies, modeling, computer simulations, etc. Flaring system is the controlled burning of natural gas in the course of routine oil and gas production operations. This burning occurs at the end of a flare stack or boom. The combustion process will release emissions of greenhouse gases such as NO2, CO2, SO2, etc. This condition will affect the air and environment around the industrial area. Therefore, we need a simulation to create the pattern of the dissemination of pollutant. This research paper has being made to see trends in gas flaring model and current developments to predict dominant variable which gives impact to dissemination of pollutant. Fluent models used to simulate the distribution of pollutant gas coming out of the stack. While WRF model output is used to overcome the limitations of the analysis of meteorological data and atmospheric conditions in the study area. This study condition focused on transition season in 2012 at Minas area. The goal of the simulation is looking for the exact time which is most influence towards dissemination of pollutants. The most influence factor divided into two main subjects. It is the quickest wind and the slowest wind. According to the simulation results, it can be seen that quickest wind moves to horizontal way and slowest wind moves to vertical way.

Keywords: flaring system, fluent model, dissemination of pollutant, transition season

Procedia PDF Downloads 363

Authors: Somayeh Komeylian

Abstract:

The direction of arrival (DoA) estimation is the crucial aspect of the radar technologies for detecting and dividing several signal sources. In this scenario, the antenna array output modeling involves numerous parameters including noise samples, signal waveform, signal directions, signal number, and signal to noise ratio (SNR), and thereby the methods of the DoA estimation rely heavily on the generalization characteristic for establishing a large number of the training data sets. Hence, we have analogously represented the two different optimization models of the DoA estimation; (1) the implementation of the decision directed acyclic graph (DDAG) for the multiclass least-squares support vector machine (LS-SVM), and (2) the optimization method of the deep neural network (DNN) radial basis function (RBF). We have rigorously verified that the LS-SVM DDAG algorithm is capable of accurately classifying DoAs for the three classes. However, the accuracy and robustness of the DoA estimation are still highly sensitive to technological imperfections of the antenna arrays such as non-ideal array design and manufacture, array implementation, mutual coupling effect, and background radiation and thereby the method may fail in representing high precision for the DoA estimation. Therefore, this work has a further contribution on developing the DNN-RBF model for the DoA estimation for overcoming the limitations of the non-parametric and data-driven methods in terms of array imperfection and generalization. The numerical results of implementing the DNN-RBF model have confirmed the better performance of the DoA estimation compared with the LS-SVM algorithm. Consequently, we have analogously evaluated the performance of utilizing the two aforementioned optimization methods for the DoA estimation using the concept of the mean squared error (MSE).

Keywords: DoA estimation, Adaptive antenna array, Deep Neural Network, LS-SVM optimization model, Radial basis function, and MSE

Procedia PDF Downloads 79

Authors: Anika Chebrolu

Abstract:

Mono-targeted drugs can be of limited efficacy against complex diseases. Recently, multi-target drug design has been approached as a promising tool to fight against these challenging diseases. However, the scope of current computational approaches for multi-target drug design is limited. DeepLig presents a de-novo drug discovery platform that uses reinforcement learning to generate and optimize novel, potent, and multitargeted drug candidates against protein targets. DeepLig’s model consists of two networks in interplay: a generative network and a predictive network. The generative network, a Stack- Augmented Recurrent Neural Network, utilizes a stack memory unit to remember and recognize molecular patterns when generating novel ligands from scratch. The generative network passes each newly created ligand to the predictive network, which then uses multiple Graph Attention Networks simultaneously to forecast the average binding affinity of the generated ligand towards multiple target proteins. With each iteration, given feedback from the predictive network, the generative network learns to optimize itself to create molecules with a higher average binding affinity towards multiple proteins. DeepLig was evaluated based on its ability to generate multi-target ligands against two distinct proteins, multi-target ligands against three distinct proteins, and multi-target ligands against two distinct binding pockets on the same protein. With each test case, DeepLig was able to create a library of valid, synthetically accessible, and novel molecules with optimal and equipotent binding energies. We propose that DeepLig provides an effective approach to design multi-targeted drug therapies that can potentially show higher success rates during in-vitro trials.

Keywords: drug design, multitargeticity, de-novo, reinforcement learning

Procedia PDF Downloads 68

Authors: Yamin Sayyari

Abstract:

In this paper, we generalized the Jensen's inequality for m-convex functions and also we present a correction of Jensen's inequality which is a better than the generalization of this inequality for m-convex functions. Finally, we have found new lower and new upper bounds for Jensen's discrete inequality.

Keywords: Jensen's inequality, m-convex function, Convex function, Inequality

Procedia PDF Downloads 132

Authors: Basant Kumar Singh, S. Subhachandhar, Vineet Ranjan Tripathi, Amit Kumar Singh, Uttam Singh, Santosh Kumar Lal

Abstract:

Productivity of Blast Furnace is largely impacted by fuel efficiency and controlling heat loss is one of the enabling parameters for achieving lower fuel rate. 'I' Blast Furnace is the latest and largest Blast Furnace of Tata Steel Jamshedpur with working volume of 3230 m³ and with rated capacity of 3.055 million tons per annum. Optimizing heat losses in Belly and Bosh zone remained major challenge for blast furnace operators after its commissioning. 'I' Blast has installed Cast Iron & Copper Staves cooling members where copper staves are installed in Belly, Bosh & Lower Stack whereas cast iron staves are installed in upper stack area. Stave cooled Blast Furnaces are prone to higher heat losses in Belly and Bosh region with an increase in coal injection rate as Bosh gas volume increases. Under these conditions, managing gas flow pattern through proper burden distribution, casting techniques & by maintaining desired raw material qualities are of utmost importance for sustaining high injection rates. This study details, the burden distribution control by Ore & Coke ratio adjustment at wall and center of Blast Furnace as the coal injection rates increased from 140 kg/thm to 210 kg/thm. Control of blowing parameters, casting philosophy, specification for raw materials & devising operational practice for controlling heat losses is also elaborated with the model that is used to visualize heat loss pattern in different zones of Blast Furnace.

Keywords: blast furnace, staves, gas flow pattern, belly/bosh heat losses, ore/coke ratio, blowing parameters, casting, operation practice

Procedia PDF Downloads 337

Authors: Mohammed Husein Mohammed Rashid

Abstract:

Let Τ = U |Τ| be a polar decomposition of a bounded linear operator T on a complex Hilbert space with ker U = ker |T|. T is said to be class p-w A(s,t) if (|T*|ᵗ|T|²ˢ|T*|ᵗ )ᵗᵖ/ˢ⁺ᵗ ≥|T*|²ᵗᵖ and |T|²ˢᵖ ≥ (|T|ˢ|T*|²ᵗ|T|ˢ)ˢᵖ/ˢ⁺ᵗ with 0Keywords: class p-w A (s, t), normaloid, isoloid, finite, orthogonality

Procedia PDF Downloads 92

Authors: Donglan Zha, Jiansong Si

Abstract:

Because of its flexibility, CES attracts much interest in economic growth and programming models, and the macroeconomics or micro-macro models. This paper focuses on the development, estimating methods of CES production function considering energy as an input. We leave for future research work of relaxing the assumption of constant returns to scale, the introduction of potential input factors, and the generalization method of the optimal nested form of multi-factor production functions.

Keywords: bias of technical change, CES production function, elasticity of substitution, energy input

Procedia PDF Downloads 264

Authors: Mathew Wakefield, Matthew Mitchell, Lisa Wise, Christopher McCarthy

Abstract:

The properties of memory representations in artificial neural networks have cognitive implications. Distributed representations that encode instances as a pattern of activity across layers of nodes afford memory compression and enforce the selection of a single point in instance space. These encoding schemes also appear to distort the representational space, as well as trading off the ability to validate that input information is within the bounds of past experience. In contrast, a localist representation which encodes some meaningful information into individual nodes in a network layer affords less memory compression while retaining the integrity of the representational space. This allows the validity of an input to be determined. The validity (or familiarity) of input along with the capacity of localist representation for multiple instance selections affords a memory sampling approach that dynamically balances the bias-variance trade-off. When the input is familiar, bias may be high by referring only to the most similar instances in memory. When the input is less familiar, variance can be increased by referring to more instances that capture a broader range of features. Using this approach in a localist instance memory network, an experiment demonstrates a relationship between representational conflict, generalization performance, and memorization demand. Relatively small sampling ranges produce the best performance on a classic machine learning dataset of visual objects. Combining memory validity with conflict detection produces a reliable confidence judgement that can separate responses with high and low error rates. Confidence can also be used to signal the need for supervisory input. Using this judgement, the need for supervised learning as well as memory encoding can be substantially reduced with only a trivial detriment to classification performance.

Keywords: artificial neural networks, representation, memory, conflict monitoring, confidence

Procedia PDF Downloads 111

Authors: Helen Durgante, Mariana F. Sparremberger, Flavia C. Bernardes, Debora D. DellAglio

Abstract:

Health promotion programmes for retirees, based on Positive Psychology perspectives for the development of strengths and virtues, demand broadened empirical investigation in Brazil. In the case of evidence-based applied research, it is suggested feasibility studies are conducted prior to efficacy trials of the intervention, in order to identify and rectify possible faults in the design and implementation of the intervention. The aim of this study was to evaluate the feasibility of a multicomponent Positive Psychology programme for health promotion of retirees, based on Cognitive Behavioural Therapy and Positive Psychology perspectives. The programme structure included six weekly group sessions (two hours each) encompassing strengths such as Values and self-care, Optimism, Empathy, Gratitude, Forgiveness, and Meaning of life and work. The feasibility criteria evaluated were: Demand, Acceptability, Satisfaction with the programme and with the moderator, Comprehension/Generalization of contents, Evaluation of the moderator (Social Skills and Integrity/Fidelity), Adherence, and programme implementation. Overall, 11 retirees (F=11), age range 54-75, from the metropolitan region of Porto Alegre-RS-Brazil took part in the study. The instruments used were: Qualitative Admission Questionnaire; Moderator Field Diary; the Programme Evaluation Form to assess participants satisfaction with the programme and with the moderator (a six-item 4-point likert scale), and Comprehension/Generalization of contents (a three-item 4-point likert scale); Observers’ Evaluation Form to assess the moderator Social Skills (a five-item 4-point likert scale), Integrity/Fidelity (a 10 item 4-point likert scale), and Adherence (a nine-item 5-point likert scale). Qualitative data were analyzed using content analysis. Descriptive statistics as well as Intraclass Correlations coefficients were used for quantitative data and inter-rater reliability analysis. The results revealed high demand (N = 55 interested people) and acceptability (n = 10 concluded the programme with overall 88.3% frequency rate), satisfaction with the program and with the moderator (X = 3.76, SD = .34), and participants self-report of Comprehension/Generalization of contents provided in the programme (X = 2.82, SD = .51). In terms of the moderator Social Skills (X = 3.93; SD = .40; ICC = .752 [IC = .429-.919]), Integrity/Fidelity (X = 3.93; SD = .31; ICC = .936 [IC = .854-.981]), and participants Adherence (X = 4.90; SD = .29; ICC = .906 [IC = .783-.969]), evaluated by two independent observers present in each session of the programme, descriptive and Intraclass Correlation results were considered adequate. Structural changes were introduced in the intervention design and implementation methods, as well as the removal of items from questionnaires and evaluation forms. The obtained results were satisfactory, allowing changes to be made for further efficacy trials of the programme. Results are discussed taking cultural and contextual demands in Brazil into account.

Keywords: feasibility study, health promotion, positive psychology intervention, programme evaluation, retirees

Procedia PDF Downloads 177

Authors: Filippo Portera

Abstract:

Usually standard algorithms employ a loss where each error is the mere absolute difference between the true value and the prediction, in case of a regression task. In the present, we present several error weighting schemes that are a generalization of the consolidated routine. We study both a binary classification model for Support Vextor Classification and a regression net for Multylayer Perceptron. Results proves that the error is never worse than the standard procedure and several times it is better.

Keywords: loss, binary-classification, MLP, weights, regression

Procedia PDF Downloads 76

Authors: Daniel Li

Abstract:

We present a novel application of neural rendering methods to confocal microscopy. Neural rendering and implicit neural representations have developed at a remarkable pace, and are prevalent in modern 3D computer vision literature. However, they have not yet been applied to optical microscopy, an important imaging field where 3D volume information may be heavily sought after. In this paper, we employ neural rendering on confocal microscopy focus stack data and share the results. We highlight the benefits and potential of adding neural rendering to the toolkit of microscopy image processing techniques.

Keywords: neural rendering, implicit neural representations, confocal microscopy, medical image processing

Procedia PDF Downloads 641

Authors: Mouhamadou Dosso

Abstract:

This paper presents methods of spectral dichotomy of a matrix which compute spectral projectors on the subspace associated with the eigenvalues external to the parabolas described by a general equation. These methods are modifications of the one proposed in [A. N. Malyshev and M. Sadkane, SIAM J. MATRIX ANAL. APPL. 18 (2), 265-278, 1997] which uses the spectral dichotomy method of a matrix with respect to the imaginary axis. Theoretical and algorithmic aspects of the methods are developed. Numerical results obtained by applying methods presented on matrices are reported.

Keywords: spectral dichotomy method, spectral projector, eigensubspaces, eigenvalue

Procedia PDF Downloads 74

Authors: M. Rodionov, Z. Dedovets

Abstract:

This article discusses ways to implement a differentiated approach to developing academic motivation for mathematical studies which relies on defining the primary structural characteristics of motivation. The following characteristics are considered: features of realization of cognitive activity, meaning-making characteristics, level of generalization and consistency of knowledge acquired by personal experience. The assessment of the present level of individual student understanding of each component of academic motivation is the basis for defining the relevant educational strategy for its further development.

Keywords: learning activity, mathematics, motivation, student

Procedia PDF Downloads 405

Authors: Abdullah E. Al-Mazrooei

Abstract:

In this paper, we present a new kind of the bilinear systems in the form of state space model. The evolution of this system depends on the product of state vector by its self. The well known Lotak Volterra and Lorenz models are special cases of this new model. We also present here a generalization of Kalman filter which is suitable to work with the new bilinear model. An application to real measurements is introduced to illustrate the efficiency of the proposed algorithm.

Keywords: bilinear systems, state space model, Kalman filter, application, models

Procedia PDF Downloads 412

Authors: M.A.Bakry1, *, Aryn T. Shafeek1, +

Abstract:

In this article, absolute Parallelism geometry is used to study the torsional gravitational field. And discovered the tetrad fields, torsion vector, and torsion scalar of Schwarzschild space. The new solution of the torsional gravitational field is a generalization of Schwarzschild in the context of general relativity. The results are applied to the planetary orbits.

Keywords: absolute parallelism geometry, tetrad fields, torsion vectors, torsion scalar

Procedia PDF Downloads 127

Authors: Jingqi Li

Abstract:

A new vertical carbon nanotube field effect transistor (CNTFET) has been developed. The source, drain and gate are vertically stacked in this structure. The carbon nanotubes are put on the side wall of the vertical stack. Unique transfer characteristics which depend on both silicon type and the sign of drain voltage have been observed in silicon integrated CNTFETs. The significant advantage of this CNTFET is that the short channel of the transistor can be fabricated without using complicate lithography technique.

Keywords: carbon nanotubes, field-effect transistors, electrical property, short channel fabrication

Procedia PDF Downloads 338

Authors: Z. Gharaee

Abstract:

Recognizing actions performed by others is important in our daily lives since it is necessary for communicating with others in a proper way. We perceive an action by observing the kinematics of motions involved in the performance. We use our experience and concepts to make a correct recognition of the actions. Although building the action concepts is a life-long process, which is repeated throughout life, we are very efficient in applying our learned concepts in analyzing motions and recognizing actions. Experiments on the subjects observing the actions performed by an actor show that an action is recognized after only about two hundred milliseconds of observation. In this study, hierarchical action recognition architecture is proposed by using growing grid layers. The first-layer growing grid receives the pre-processed data of consecutive 3D postures of joint positions and applies some heuristics during the growth phase to allocate areas of the map by inserting new neurons. As a result of training the first-layer growing grid, action pattern vectors are generated by connecting the elicited activations of the learned map. The ordered vector representation layer receives action pattern vectors to create time-invariant vectors of key elicited activations. Time-invariant vectors are sent to second-layer growing grid for categorization. This grid creates the clusters representing the actions. Finally, one-layer neural network developed by a delta rule labels the action categories in the last layer. System performance has been evaluated in an experiment with the publicly available MSR-Action3D dataset. There are actions performed by using different parts of human body: Hand Clap, Two Hands Wave, Side Boxing, Bend, Forward Kick, Side Kick, Jogging, Tennis Serve, Golf Swing, Pick Up and Throw. The growing grid architecture was trained by applying several random selections of generalization test data fed to the system during on average 100 epochs for each training of the first-layer growing grid and around 75 epochs for each training of the second-layer growing grid. The average generalization test accuracy is 92.6%. A comparison analysis between the performance of growing grid architecture and self-organizing map (SOM) architecture in terms of accuracy and learning speed show that the growing grid architecture is superior to the SOM architecture in action recognition task. The SOM architecture completes learning the same dataset of actions in around 150 epochs for each training of the first-layer SOM while it takes 1200 epochs for each training of the second-layer SOM and it achieves the average recognition accuracy of 90% for generalization test data. In summary, using the growing grid network preserves the fundamental features of SOMs, such as topographic organization of neurons, lateral interactions, the abilities of unsupervised learning and representing high dimensional input space in the lower dimensional maps. The architecture also benefits from an automatic size setting mechanism resulting in higher flexibility and robustness. Moreover, by utilizing growing grids the system automatically obtains a prior knowledge of input space during the growth phase and applies this information to expand the map by inserting new neurons wherever there is high representational demand.

Keywords: action recognition, growing grid, hierarchical architecture, neural networks, system performance

Procedia PDF Downloads 142

Authors: Sabin Aslam, Sultan Habibullah Khan, James G. Thomson, Abhaya M. Dandekar

Abstract:

Losses due to viral diseases are posing a serious threat to crop production. A quick breakdown of resistance to viruses like Cotton Leaf Curl Virus (CLCuV) demands the application of a proficient technology to engineer durable resistance. Gene stacking has recently emerged as a potential approach for integrating multiple genes in crop plants. In the present study, recombinase technology has been used for site-specific gene stacking. A target vector (pG-Rec) was designed for engineering a predetermined specific site in the plant genome whereby genes can be stacked repeatedly. Using Agrobacterium-mediated transformation, the pG-Rec was transformed into Coker-312 along with Nicotiana tabacum L. cv. Xanthi and Nicotiana benthamiana. The transgene analysis of target lines was conducted through junction PCR. The transgene positive target lines were used for further transformations to site-specifically stack two genes of interest using Bxb1 and PhiC31 recombinases. In the first instance, Cas9 driven by multiplex gRNAs (for Rep gene of CLCuV) was site-specifically integrated into the target lines and determined by the junction PCR and real-time PCR. The resulting plants were subsequently used to stack the second gene of interest (AVP3 gene from Arabidopsis for enhancing cotton plant growth). The addition of the genes is simultaneously achieved with the removal of marker genes for recycling with the next round of gene stacking. Consequently, transgenic marker-free plants were produced with two genes stacked at the specific site. These transgenic plants can be potential germplasm to introduce resistance against various strains of cotton leaf curl virus (CLCuV) and abiotic stresses. The results of the research demonstrate gene stacking in crop plants, a technology that can be used to introduce multiple genes sequentially at predefined genomic sites. The current climate change scenario highlights the use of such technologies so that gigantic environmental issues can be tackled by several traits in a single step. After evaluating virus resistance in the resulting plants, the lines can be a primer to initiate stacking of further genes in Cotton for other traits as well as molecular breeding with elite cotton lines.

Keywords: cotton, CRISPR/Cas9, gene stacking, genome editing, recombinases

Procedia PDF Downloads 128

Authors: Mohsen Zahraei

Abstract:

‎In this paper, ‎the notion of ‎rank-k numerical range of rectangular complex matrix polynomials‎ ‎are introduced. ‎Some algebraic and geometrical properties are investigated. ‎Moreover, ‎for ε>0 the notion of Birkhoff-James approximate orthogonality sets for ε-higher ‎rank numerical ranges of rectangular matrix polynomials is also introduced and studied. ‎The proposed definitions yield a natural generalization of the standard higher rank numerical ranges.

Keywords: ‎‎Rank-k numerical range‎, ‎isometry‎, ‎numerical range‎, ‎rectangular matrix polynomials

Procedia PDF Downloads 435

Authors: M. Chumburidze, N. Salia, T. Namchevadze

Abstract:

This work deal with case processing problems in business. The task of strategic credit requirements management of cases portfolio is discussed. The information model of credit requirements in a binary tree diagram is considered. The algorithms to solve issues of prioritizing clusters of cases in business have been investigated. An implementation of priority queues to support case management operations has been presented. The corresponding pseudo codes for the programming application have been constructed. The tools applied in this development are based on binary tree ordering algorithms, optimization theory, and business management methods.

Keywords: credit network, case portfolio, binary tree, priority queue, stack

Procedia PDF Downloads 122

Authors: A. Giniatoulline

Abstract:

A nonlinear model of the mathematical fluid dynamics which describes the motion of an incompressible viscous rotating fluid in a homogeneous gravitational field is considered. The model is a generalization of the known Navier-Stokes system with the addition of the Coriolis parameter and the equations for changeable density. An explicit algorithm for the solution is constructed, and the proof of the existence and uniqueness theorems for the strong solution of the nonlinear problem is given. For the linear case, the localization and the structure of the spectrum of inner waves are also investigated.

Keywords: Galerkin method, Navier-Stokes equations, nonlinear partial differential equations, Sobolev spaces, stratified fluid

Procedia PDF Downloads 289

Authors: Khaled I. Nawafleh

Abstract:

In this work, we apply the method of Hamilton-Jacobi to obtain solutions of Hamiltonian systems in classical mechanics with two certain structures: the first structure plays a central role in the theory of time-dependent Hamiltonians, whilst the second is used to treat classical Hamiltonians, including dissipation terms. It is proved that the generalization of problems from the calculus of variation methods in the nonstationary case can be obtained naturally in Hamilton-Jacobi formalism. Then, another expression of geometry of the Hamilton Jacobi equation is retrieved for Hamiltonians with time-dependent and frictional terms. Both approaches shall be applied to many physical examples.

Keywords: Hamilton-Jacobi, time dependent lagrangians, dissipative systems, variational principle

Procedia PDF Downloads 150

Authors: Alona Faktor

Abstract:

In this work, we present a NN-based computational model that can perform attention shifts according to high-level instruction. The instruction specifies the type of attentional shift using explicit geometrical relation. The instruction also can be of cognitive nature, specifying more complex human-human interaction or human-object interaction, or object-object interaction. Applying this approach sequentially allows obtaining a structural description of an image. A novel data-set of interacting humans and objects is constructed using a computer graphics engine. Using this data, we perform systematic research of relational segmentation shifts.

Keywords: cognitive science, attentin, deep learning, generalization

Procedia PDF Downloads 178

Authors: Arpit Rai

Abstract:

In this work, we propose PatchMix, a semi-supervised method for pre-training visual representations. PatchMix mixes patches of two images and then solves an auxiliary task of predicting the label of each patch in the mixed image. Our experiments on the CIFAR-10, 100 and the SVHN dataset show that the representations learned by this method encodes useful information for transfer to new tasks and outperform the baseline Residual Network encoders by on CIFAR 10 by 12% on ResNet 101 and 2% on ResNet-56, by 4% on CIFAR-100 on ResNet101 and by 6% on SVHN dataset on the ResNet-101 baseline model.

Keywords: self-supervised learning, representation learning, computer vision, generalization

Procedia PDF Downloads 69