Search results for: convergence

Authors: Xueqing Ivy Chen, Christo De Coning

Abstract:

It is clear from practice that a gap exists between Result-Based Monitoring and Evaluation (RBME) as a discipline, and Knowledge Management (KM) on the other hand. Whereas various government departments have institutionalised these functions, KM and M&E has functioned in isolation from each other in a practical sense in the public sector. It’s therefore necessary to explore the relationship between KM and M&E and the necessity for integration, so that a convergence of these disciplines can be established. An integration of KM and M&E will lead to integration and improvement of evidence-based information and policy decision-making. M&E and KM process models are available but the complementarity between specific process steps of these process models are not exploited. A need exists to clarify the relationships between these functions in order to ensure evidence based information and policy decision-making. This paper will depart from the well-known policy process models, such as the generic model and consider recent on the interface between policy, M&E and KM.

Keywords: result-based monitoring and evaluation, RBME, knowledge management, KM, evident based decision making, public policy, information systems, institutional arrangement

Procedia PDF Downloads 117

Authors: Sergey Ermolin, Olga Ermolin

Abstract:

A Deep Convolution Neural Network with Triplet Loss is used to identify duplicate images in real-estate advertisements in the presence of image artifacts such as watermarking, cropping, hue/brightness adjustment, and others. The effects of batch normalization, spatial dropout, and various convergence methodologies on the resulting detection accuracy are discussed. For comparative Return-on-Investment study (per industry request), end-2-end performance is benchmarked on both Nvidia Titan GPUs and Intel’s Xeon CPUs. A new real-estate dataset from San Francisco Bay Area is used for this work. Sufficient duplicate detection accuracy is achieved to supplement other database-grounded methods of duplicate removal. The implemented method is used in a Proof-of-Concept project in the real-estate industry.

Keywords: visual recognition, convolutional neural networks, triplet loss, spatial batch normalization with dropout, duplicate removal, advertisement technologies, performance benchmarking

Procedia PDF Downloads 300

Authors: Abdullah M. Almeshal, Mohammad R. Alenezi, Muhammad Moaz

Abstract:

In this paper, we discuss the performance of applying hybrid spiral dynamic bacterial chemotaxis (HSDBC) optimisation algorithm on an intelligent controller for a differential drive robot. A unicycle class of differential drive robot is utilised to serve as a basis application to evaluate the performance of the HSDBC algorithm. A hybrid fuzzy logic controller is developed and implemented for the unicycle robot to follow a predefined trajectory. Trajectories of various frictional profiles and levels were simulated to evaluate the performance of the robot at different operating conditions. Controller gains and scaling factors were optimised using HSDBC and the performance is evaluated in comparison to previously adopted optimisation algorithms. The HSDBC has proven its feasibility in achieving a faster convergence toward the optimal gains and resulted in a superior performance.

Keywords: differential drive robot, hybrid fuzzy controller, optimization, path tracking, unicycle robot

Procedia PDF Downloads 430

Authors: Changqiao Wu, Guoqing Ding, Xin Chen

Abstract:

In this paper, ways of modeling dynamic measurement systems are discussed. Specially, for linear system with single-input single-output, it could be modeled with shallow neural network. Then, gradient based optimization algorithms are used for searching the proper coefficients. Besides, method with normal equation and second order gradient descent are proposed to accelerate the modeling process, and ways of better gradient estimation are discussed. It shows that the mathematical essence of the learning objective is maximum likelihood with noises under Gaussian distribution. For conventional gradient descent, the mini-batch learning and gradient with momentum contribute to faster convergence and enhance model ability. Lastly, experimental results proved the effectiveness of second order gradient descent algorithm, and indicated that optimization with normal equation was the most suitable for linear dynamic models.

Keywords: dynamic system modeling, neural network, normal equation, second order gradient descent

Procedia PDF Downloads 87

Authors: Dao Phuong Nam, Tran Van Tuyen, Do Trong Tan, Bui Minh Dinh, Nguyen Van Huong

Abstract:

In this paper, we investigate the adaptive optimal control law for continuous-time systems with input disturbances and unknown parameters. This paper extends previous works to obtain the robust control law of uncertain systems. Through theoretical analysis, an adaptive dynamic programming (ADP) based optimal control is proposed to stabilize the closed-loop system and ensure the convergence properties of proposed iterative algorithm. Moreover, the global asymptotic stability (GAS) for closed system is also analyzed. The theoretical analysis for continuous-time systems and simulation results demonstrate the performance of the proposed algorithm for an inverted pendulum system.

Keywords: approximate/adaptive dynamic programming, ADP, adaptive optimal control law, input state stability, ISS, inverted pendulum

Procedia PDF Downloads 162

Authors: Aritras Roy, Rinku Mukherjee

Abstract:

The current study couples a quasi-steady Vortex Lattice Method and a camber correcting technique, ‘Decambering’ for unsteady post-stall flow prediction. The wake is force-free and discrete such that the wake lattices move with the free-stream once shed from the wing. It is observed that the time-averaged unsteady coefficient of lift sees a relative drop at post-stall angles of attack in comparison to its steady counterpart for some angles of attack. Multiple solutions occur at post-stall and three different algorithms to choose solutions in these regimes show both unsteadiness and non-convergence of the iterations. The distribution of coefficient of lift on the wing span also shows sawtooth. Distribution of vorticity changes both along span and in the direction of the free-stream as the wake develops over time with distinct roll-up, which increases with time.

Keywords: post-stall, unsteady, wing, aerodynamics

Procedia PDF Downloads 339

Authors: Pyung Soo Kim, Eung Hyuk Lee, Mun Suck Jang

Abstract:

A packet loss probability (PLP) estimation filter with finite memory structure is proposed to estimate the packet rate mean and variance of the input traffic process in real-time while removing undesired system and measurement noises. The proposed PLP estimation filter is developed under a weighted least square criterion using only the finite traffic measurements on the most recent window. The proposed PLP estimation filter is shown to have several inherent properties such as unbiasedness, deadbeat, robustness. A guideline for choosing appropriate window length is described since it can affect significantly the estimation performance. Using computer simulations, the proposed PLP estimation filter is shown to be superior to the Kalman filter for the temporarily uncertain system. One possible explanation for this is that the proposed PLP estimation filter can have greater convergence time of a filtered estimate as the window length M decreases.

Keywords: packet loss probability estimation, finite memory filter, infinite memory filter, Kalman filter

Procedia PDF Downloads 631

Authors: Chandan Pramanik, Bikramjit Chanda

Abstract:

Large metal mines operated with open-cast mining methods must transition to underground mining at the conclusion of the operation; however, this requires a period of a difficult time when production convergence due to interference between the two mining methods. A transition model with collaborative mining operations is presented and established in this work, based on the case of the South Kaliapani Underground Project, to address these technical issues of inadequate production security and other mining challenges during the transition phase and beyond. By integrating the technology of the small-scale Drift and Fill method and Highly productive Sub Level Open Stoping at deep section, this hybrid mining concept tries to eliminate major bottlenecks and offers an optimized production profile with the safe and sustainable operation. Considering every geo-mining aspect, this study offers a genuine and precise technical deliberation for the transition from open pit to underground mining.

Keywords: drift and fill, geo-mining aspect, sublevel open stoping, underground mining method

Procedia PDF Downloads 62

Authors: Bawadi M. A., Abbad J. A., Baras E. A.

Abstract:

This study was conducted to assess wind energy resources in the area of Al-Adwas Ras-Huwirah Hadhramout Governorate, Yemen, through using statistical calculations, the Weibull model and SPSS program were used in the monthly and the annual to analyze the wind energy resource; the convergence of wind energy; turbine efficiency in the selected area. Wind speed data was obtained from NASA over a period of ten years (2010-2019) and at heights of 50 m above ground level. Probability distributions derived from wind data and their distribution parameters are determined. The density probability function is fitted to the measured probability distributions on an annual basis. This study also involves locating preliminary sites for wind farms using Geographic Information System (GIS) technology. This further leads to maximizing the output energy from the most suitable wind turbines in the proposed site.

Keywords: wind speed analysis, Yemen wind energy, wind power density, Weibull distribution model

Procedia PDF Downloads 49

Authors: Ruchi, A. M. Acu, Purshottam N. Agrawal

Abstract:

The purpose of this paper to introduce the Durrmeyer type modification of q-generalized-Bernstein operators which include the Bernstein polynomials in the particular α = 0. We investigate the rate of convergence by means of the Lipschitz class and the Peetre’s K-functional. Also, we define the bivariate case of Durrmeyer type modification of q-generalized-Bernstein operators and study the degree of approximation with the aid of the partial modulus of continuity and the Peetre’s K-functional. Finally, we introduce the GBS (Generalized Boolean Sum) of the Durrmeyer type modification of q- generalized-Bernstein operators and investigate the approximation of the Bögel continuous and Bögel differentiable functions with the aid of the Lipschitz class and the mixed modulus of smoothness.

Keywords: Bögel continuous, Bögel differentiable, generalized Boolean sum, Peetre’s K-functional, Lipschitz class, mixed modulus of smoothness

Procedia PDF Downloads 163

Authors: Liou Shyhnan, Chia Han Yang

Abstract:

The study aims to investigate the promotion of innovation in the development of cultural and creative industries (CCI) and apply research on culture and creativity to this promotion. Using the research perspectives of culture and creativity as the starting points, this study has examined the challenges, trends, and opportunities that have emerged from the development of the CCI until the present. It is found that a definite context of cause and effect exist between them, and that a homologous theoretical basis can be used to understand and interpret them. Based on the characteristics of the aforementioned challenges and trends, this study has compiled two main theoretical systems for conducting research on culture and creativity: (i) reciprocal process between creativity and culture, and (ii) a mechanism for innovation involving multicultural convergence. Both theoretical systems were then used as the foundation to arrive at possible research propositions relating to the two developmental systems. This was respectively done through identification of the theoretical context through a literature review, and interviews and observations of actual case studies within Taiwan’s CCI. In so doing, the critical factors that can address the aforementioned challenges and trends were discovered. Our results indicated that, for reciprocal process between creativity and culture, we recognize that culture serves as creative resources in cultural and creative industries. According to shared consensus, culture provides symbolic meanings and emotional attachment for products and experiences offered by CCI. Besides, different cultures vary in their effects on creativity processes and standards, thus engendering distinctive preferences for and evaluations of the creative expressions and experiences of CCIs. In addition, we identify that creativity serves as the engine for driving the continuation and rebirth of cultures. Accounting for the core of culture, the employment of technology, design, and business facilitates the transformation and innovation mechanism for promoting culture continuity. In addition, with cultural centered, the digital technology, design thinking, and business model are critical constitutes of the innovation mechanism to promote the cultural continuity. Regarding cultural preservation and regeneration of local spaces and folk customs, we argue that the preservation and regeneration of local spaces and cultural cultures must embody the interactive experiences of present-day life. And cultural space and folk custom would regenerate with interact and experience in modern life. Regarding innovation mechanism for multicultural convergence, we propose that innovative stakeholders from different disciplines (e.g., creators, designers, engineers, and marketers) in CCIs rely on the establishment of a cocreation mechanism to promote interdisciplinary interaction. Furthermore, CCI development needs to develop a cocreation mechanism for enhancing the interdisciplinary collaboration among CCI innovation stakeholders. We further argue multicultural mixing would enhance innovation in developing CCI, and assuming an open and mutually enlightening attitude to enrich one another’s cultures in the multicultural exchanges under globalization will create diversity in homogenous CCIs. Finally, for promoting innovation in developing cultural and creative industries, we further propose a model for joint knowledge creation that can be established for enhancing the mutual reinforcement of theoretical and practical research on culture and creativity.

Keywords: culture and creativity, innovation, cultural and creative industries, cultural mixing

Procedia PDF Downloads 290

Authors: Riadh Zorgati, Thomas Triboulet

Abstract:

In quite diverse application areas such as astronomy, medical imaging, geophysics or nondestructive evaluation, many problems related to calibration, fitting or estimation of a large number of input parameters of a model from a small amount of output noisy data, can be cast as inverse problems. Due to noisy data corruption, insufficient data and model errors, most inverse problems are ill-posed in a Hadamard sense, i.e. existence, uniqueness and stability of the solution are not guaranteed. A wide class of inverse problems in physics relates to the Fredholm equation of the first kind. The ill-posedness of such inverse problem results, after discretization, in a very ill-conditioned linear system of equations, the condition number of the associated matrix can typically range from 109 to 1018. This condition number plays the role of an amplifier of uncertainties on data during inversion and then, renders the inverse problem difficult to handle numerically. Similar problems appear in other areas such as numerical optimization when using interior points algorithms for solving linear programs leads to face ill-conditioned systems of linear equations. Devising efficient solution approaches for such system of equations is therefore of great practical interest. Efficient iterative algorithms are proposed for solving a system of linear equations. The approach is based on a preconditioning of the initial matrix of the system with an approximation of a generalized inverse leading to a stochastic preconditioned matrix. This approach, valid for non-negative matrices, is first extended to hermitian, semi-definite positive matrices and then generalized to any complex rectangular matrices. The main results obtained are as follows: 1) We are able to build a generalized inverse of any complex rectangular matrix which satisfies the convergence condition requested in iterative algorithms for solving a system of linear equations. This completes the (short) list of generalized inverse having this property, after Kaczmarz and Cimmino matrices. Theoretical results on both the characterization of the type of generalized inverse obtained and the convergence are derived. 2) Thanks to its properties, this matrix can be efficiently used in different solving schemes as Richardson-Tanabe or preconditioned conjugate gradients. 3) By using Lp norms, we propose generalized Kaczmarz’s type matrices. We also show how Cimmino's matrix can be considered as a particular case consisting in choosing the Euclidian norm in an asymmetrical structure. 4) Regarding numerical results obtained on some pathological well-known test-cases (Hilbert, Nakasaka, …), some of the proposed algorithms are empirically shown to be more efficient on ill-conditioned problems and more robust to error propagation than the known classical techniques we have tested (Gauss, Moore-Penrose inverse, minimum residue, conjugate gradients, Kaczmarz, Cimmino). We end on a very early prospective application of our approach based on stochastic matrices aiming at computing some parameters (such as the extreme values, the mean, the variance, …) of the solution of a linear system prior to its resolution. Such an approach, if it were to be efficient, would be a source of information on the solution of a system of linear equations.

Keywords: conditioning, generalized inverse, linear system, norms, stochastic matrix

Procedia PDF Downloads 104

Authors: Oğuzhan Hasançebi, Saeid Kazemzadeh Azad

Abstract:

Keywords: structural design optimization, optimal sizing, metaheuristics, self-adaptive step-size search, steel trusses, steel frames

Procedia PDF Downloads 334

Authors: V. Tabrizi, A. Vali, R. GHasemi, V. Behnamgol

Abstract:

In this article, a nonlinear model of an under actuated six degrees of freedom (6 DOF) quadrotor UAV is derived on the basis of the Newton-Euler formula. The derivation comprises determining equations of the motion of the quadrotor in three dimensions and approximating the actuation forces through the modeling of aerodynamic coefficients and electric motor dynamics. The robust nonlinear control strategy includes a smooth second order non-singular terminal sliding mode control which is applied to stabilizing this model. The control method is on the basis of super twisting algorithm for removing the chattering and producing smooth control signal. Also, nonsingular terminal sliding mode idea is used for introducing a nonlinear sliding variable that guarantees the finite time convergence in sliding phase. Simulation results show that the proposed algorithm is robust against uncertainty or disturbance and guarantees a fast and precise control signal.

Keywords: quadrotor UAV, nonsingular terminal sliding mode, second order sliding mode t, electronics, control, signal processing

Procedia PDF Downloads 404

Authors: Shubham Jaiswal

Abstract:

During modeling of transport phenomena in porous media, many nonlinear partial differential equations (NPDEs) encountered which greatly described the convection, diffusion and reaction process. To solve such types of nonlinear problems, a reliable and efficient technique is needed. In this article, the numerical solution of NPDEs encountered in porous media is derived. Here Jacobi collocation method is used to solve the considered problems which convert the NPDEs in systems of nonlinear algebraic equations that can be solved using Newton-Raphson method. The numerical results of some illustrative examples are reported to show the efficiency and high accuracy of the proposed approach. The comparison of the numerical results with the existing analytical results already reported in the literature and the error analysis for each example exhibited through graphs and tables confirms the exponential convergence rate of the proposed method.

Keywords: nonlinear porous media equation, shifted Jacobi polynomials, operational matrix, spectral collocation method

Procedia PDF Downloads 402

Authors: Nihal Murali, Kunal Gupta, Surekha Bhanot

Abstract:

Training of artificial neural networks (ANNs) using reinforcement learning (RL) techniques is being widely discussed in the robot learning literature. The high model complexity of ANNs along with the model-free nature of RL algorithms provides a desirable combination for many robotics applications. There is a huge need for algorithms that generalize using raw sensory inputs, such as vision, without any hand-engineered features or domain heuristics. In this paper, the standard control problem of line following robot was used as a test-bed, and an ANN controller for the robot was trained on images from a live video feed using Q-learning. A virtual agent was first trained in simulation environment and then deployed onto a robot’s hardware. The robot successfully learns to traverse a wide range of curves and displays excellent generalization ability. Qualitative analysis of the evolution of policies, performance and weights of the network provide insights into the nature and convergence of the learning algorithm.

Keywords: artificial neural networks, q-learning, reinforcement learning, robot learning

Procedia PDF Downloads 342

Authors: Thinh Ngo, Paul Rad, Brian Kelley

Abstract:

Noise estimation is essential in today wireless systems for power control, adaptive modulation, interference suppression and quality of service. Deep learning (DL) has already been applied in the physical layer for modulation and signal classifications. Unacceptably low accuracy of less than 50% is found to undermine traditional application of DL classification for SNR prediction. In this paper, we use divide-and-conquer algorithm and classifier fusion method to simplify SNR classification and therefore enhances DL learning and prediction. Specifically, multiple CNNs are used for classification rather than a single CNN. Each CNN performs a binary classification of a single SNR with two labels: less than, greater than or equal. Together, multiple CNNs are combined to effectively classify over a range of SNR values from −20 ≤ SNR ≤ 32 dB.We use pre-trained CNNs to predict SNR over a wide range of joint channel parameters including multiple Doppler shifts (0, 60, 120 Hz), power-delay profiles, and signal-modulation types (QPSK,16QAM,64-QAM). The approach achieves individual SNR prediction accuracy of 92%, composite accuracy of 70% and prediction convergence one order of magnitude faster than that of traditional estimation.

Keywords: classification, CNN, deep learning, prediction, SNR

Procedia PDF Downloads 94

Authors: Anusuya Ghosh, Vishnu Narayanan

Abstract:

The approximation of convex set by semidefinite representable set plays an important role in semidefinite programming, especially in modern convex optimization. To optimize a linear function over a convex set is a hard problem. But optimizing the linear function over the semidefinite representable set which approximates the convex set is easy to solve as there exists numerous efficient algorithms to solve semidefinite programming problems. So, our approximation technique is significant in optimization. We develop a technique to approximate any closed convex set, say K by compactly semidefinite representable set. Further we prove that there exists a sequence of compactly semidefinite representable sets which give tighter approximation of the closed convex set, K gradually. We discuss about the convergence of the sequence of compactly semidefinite representable sets to closed convex set K. The recession cone of K and the recession cone of the compactly semidefinite representable set are equal. So, we say that the sequence of compactly semidefinite representable sets converge strongly to the closed convex set. Thus, this approximation technique is very useful development in semidefinite programming.

Keywords: semidefinite programming, semidefinite representable set, compactly semidefinite representable set, approximation

Procedia PDF Downloads 347

Authors: Seongjun Kim, Sanghoon Shim, Jinwooung Kim, Jaehwan Jung, Sung-Ah Kim

Abstract:

As the development of information and communication technology, the convergence of machine learning of the ICT area and design is attempted. In this way, it is possible to grasp the correlation between various design elements, which was difficult to grasp, and to reflect this in the design result. In architecture, there is an attempt to predict the performance, which is difficult to grasp in the past, by finding the correlation among multiple factors mainly through machine learning. In architectural design area, some attempts to predict the performance affected by various factors have been tried. With machine learning, it is possible to quickly predict performance. The aim of this study is to propose a model that predicts performance according to the block arrangement of apartment housing through machine learning and the design alternative which satisfies the performance such as the daylight hours in the most similar form to the alternative proposed by the designer. Through this study, a designer can proceed with the design considering various design alternatives and accurate performances quickly from the early design stage.

Keywords: apartment housing, machine learning, multi-objective optimization, performance prediction

Procedia PDF Downloads 437

Authors: Samia Sadouki Chibani, Abdelkamel Tari

Abstract:

Web service composition combines available services to provide new functionality. Given the number of available services with similar functionalities and different non functional aspects (QoS), the problem of finding a QoS-optimal web service composition is considered as an optimization problem belonging to NP-hard class. Thus, an optimal solution cannot be found by exact algorithms within a reasonable time. In this paper, a meta-heuristic bio-inspired is presented to address the QoS aware web service composition; it is based on Elephant Herding Optimization (EHO) algorithm, which is inspired by the herding behavior of elephant group. EHO is characterized by a process of dividing and combining the population to sub populations (clan); this process allows the exchange of information between local searches to move toward a global optimum. However, with Applying others evolutionary algorithms the problem of early stagnancy in a local optimum cannot be avoided. Compared with PSO, the results of experimental evaluation show that our proposition significantly outperforms the existing algorithm with better performance of the fitness value and a fast convergence.

Keywords: bio-inspired algorithms, elephant herding optimization, QoS optimization, web service composition

Procedia PDF Downloads 289

Authors: Md Al Masum Bhuiyan, Maria C. Mariani, Osei K. Tweneboah

Abstract:

In the present work, we develop a technique for estimating the volatility of financial time series by using stochastic differential equation. Taking the daily closing prices from developed and emergent stock markets as the basis, we argue that the incorporation of stochastic volatility into the time-varying parameter estimation significantly improves the forecasting performance via Maximum Likelihood Estimation. While using the technique, we see the long-memory behavior of data sets and one-step-ahead-predicted log-volatility with ±2 standard errors despite the variation of the observed noise from a Normal mixture distribution, because the financial data studied is not fully Gaussian. Also, the Ornstein-Uhlenbeck process followed in this work simulates well the financial time series, which aligns our estimation algorithm with large data sets due to the fact that this algorithm has good convergence properties.

Keywords: financial time series, maximum likelihood estimation, Ornstein-Uhlenbeck type models, stochastic volatility model

Procedia PDF Downloads 205

Authors: Alaa Abdullah Altaee

Abstract:

This paper presents an adaptive decision feedback equalizer (ADFE) for multi-Gbps serial links utilizing a fix-step error signal extracted from cross-points of received data symbols. The extracted signal is generated based on violation of received data symbols with minimum detection requirements at the clock and data recovery (CDR) stage. The iterations of the adaptation process search for the optimum feedback tap coefficients to maximize the data eye-opening and minimize the adaptation convergence time. The effectiveness of the proposed architecture is validated using the simulation results of a serial link designed in an IBM 130 nm 1.2V CMOS technology. The data link with variable channel lengths is analyzed using Spectre from Cadence Design Systems with BSIM4 device models.

Keywords: adaptive DFE, CMOS equalizer, error detection, serial links, timing jitter, wire-line communication

Procedia PDF Downloads 82

Authors: Seyed Reza Mirdehghan, Mohammad Reza Haeri Yazdi

Abstract:

In this paper, the stability and the performance of a two-degree-of-freedom robot under an interaction with a unknown environment has been investigated. The time when the robot returns to its initial position after an interaction and the primary resistance of the robot against the impact must be reduced. Thus, the applied torque on the motor will be reduced. The impedance control is an appropriate method for robot control in these conditions. The stability of the robot at interaction moment was transformed to be a robust stability problem. The dynamic of the unknown environment was modeled as a weight function and the stability of the robot under an interaction with the environment has been investigated using the robust control concept. To improve the performance of the system, a force controller has been designed which the normalized impedance after interaction has been reduced. The resistance of the robot has been considered as a normalized cost function and its value was 0.593. The results has showed reduction of resistance of the robot against impact and the reduction of convergence time by lower than one second.

Keywords: impedance control, control system, robots, interaction

Procedia PDF Downloads 388

Authors: Clarinda Vitorino Nhangumbe, Ercília Sousa

Abstract:

Weather derivatives are financial products used to cover non catastrophic weather events with a weather index as the underlying asset. The rainfall weather derivative pricing model is modeled based in the assumption that the rainfall dynamics follows Ornstein-Uhlenbeck process, and the partial differential equation approach is used to derive the convection-diffusion two dimensional time dependent partial differential equation, where the spatial variables are the rainfall index and rainfall depth. To compute the approximation solutions of the partial differential equation, the appropriate boundary conditions are suggested, and an explicit numerical method is proposed in order to deal efficiently with the different choices of the coefficients involved in the equation. Being an explicit numerical method, it will be conditionally stable, then the stability region of the numerical method and the order of convergence are discussed. The model is tested for real precipitation data.

Keywords: finite differences method, ornstein-uhlenbeck process, partial differential equations approach, rainfall derivatives

Procedia PDF Downloads 59

Authors: Ishay Wolf

Abstract:

In this study, the financial position of pension actors in the market during the pension system transition toward a more funded capitalized scheme is explored, mainly via an option benefit model. This is enabled by not considering the economy as a single earning cohort. We analytically demonstrate a socio-economic anomaly in the funded pension system, which is in favor of high earning cohorts on at the expense of low earning cohorts. This anomaly is realized by a lack of insurance and exposure to financial and systemic risks. Furthermore, the anomaly might lead to pension re-reform back to unfunded scheme, mostly due to political pressure. We find that a minimum pension guarantee is a rebalance mechanism to this anomaly, which increases the probability to of the sustainable pension scheme. Specifically, we argue that implementing the guarantee with an intra-generational, risk-sharing mechanism is the most efficient way to reduce the effect of this abnormality. Moreover, we exhibit the convergence process toward implementing minimum pension guarantee in many countries which have capitalized their pension systems during the last three decades, particularly among Latin America and CEE countries.

Keywords: benefits, pension scheme, put option, social security

Procedia PDF Downloads 90

Authors: Hiba Mezaache

Abstract:

The study aims to assess the financial and social efficiency of microfinance institutions in india for the period 2015-2019 by using two models of economies of scale and choosing the output direction of the data envelope analysis (DEA) method and using the MIX MARKET database. The study concluded that microfinance institutions focus on achieving financial efficiency beyond their focus on achieving social efficiency to ensure their continuity in the market. Convergence in the efficiency ratios that have been achieved, but the optimum ratios have been achieved under the changing economies of scale; Efficiency is affected by the depth of reaching low-income groups, as serving this group raises costs and risks. The importance of lending to women in rural areas and raising their awareness to ensure their financial and social empowerment; Make improvements in operating expenses, asset management, and loan personnel control in order to maximize output.

Keywords: microfinance, financial efficiency, social efficiency, mix market, microfinance institutions

Procedia PDF Downloads 117

Authors: Roman Klas, František Pochylý, Pavel Rudolf

Abstract:

This paper is focused on the CFD simulation of the radiaxial pump (i.e. mixed flow pump) with the aim to detect the reasons of Y-Q characteristic instability. The main reasons of pressure pulsations were detected by means of the analysis of velocity and pressure fields within the pump combined with the theoretical approach. Consequently, the modifications of spiral case and pump suction area were made based on the knowledge of flow conditions and the shape of dissipation function. The primary design of pump geometry was created as the base model serving for the comparison of individual modification influences. The basic experimental data are available for this geometry. This approach replaced the more complicated and with respect to convergence of all computational tasks more difficult calculation for the compressible liquid flow. The modification of primary pump consisted in inserting the three fins types. Subsequently, the evaluation of pressure pulsations, specific energy curves and visualization of velocity fields were chosen as the criterion for successful design.

Keywords: CFD, radiaxial pump, spiral case, stability

Procedia PDF Downloads 358

Authors: Santhosh A. K.

Abstract:

This work presents an efficient monolithic finite element strategy for solving thermo-fluid-structure interaction problems involving compressible fluids and linear-elastic structure. This formulation uses displacement variables for structure and velocity variables for the fluid, with no additional variables required to ensure traction, velocity, temperature, and heat flux continuity at the fluid-structure interface. Rate of convergence in each time step is quadratic, which is achieved in this formulation by deriving an exact tangent stiffness matrix. The robustness and good performance of the method is ascertained by applying the proposed strategy on a wide spectrum of problems taken from the literature pertaining to steady, transient, two dimensional, axisymmetric, and three dimensional fluid flow and conjugate heat transfer. It is shown that the current formulation gives excellent results on all the case studies conducted, which includes problems involving compressibility effects as well as problems where fluid can be treated as incompressible.

Keywords: linear thermoelasticity, compressible flow, conjugate heat transfer, monolithic FEM

Procedia PDF Downloads 161

Authors: Armin Nabaei, M. Omair Ahmad, M. N. S. Swamy

Abstract:

Many facial expression and emotion recognition methods in the traditional approaches of using LDA, PCA, and EBGM have been proposed. In recent years deep learning models have provided a unique platform addressing by automatically extracting the features for the detection of facial expression and emotions. However, deep networks require large training datasets to extract automatic features effectively. In this work, we propose an efficient emotion detection algorithm using face images when only small datasets are available for training. We design a deep network whose feature extraction capability is enhanced by utilizing several parallel modules between the input and output of the network, each focusing on the extraction of different types of coarse features with fined grained details to break the symmetry of produced information. In fact, we leverage long range dependencies, which is one of the main drawback of CNNs. We develop this work by introducing a Dynamic Soft-Margin SoftMax.The conventional SoftMax suffers from reaching to gold labels very soon, which take the model to over-fitting. Because it’s not able to determine adequately discriminant feature vectors for some variant class labels. We reduced the risk of over-fitting by using a dynamic shape of input tensor instead of static in SoftMax layer with specifying a desired Soft- Margin. In fact, it acts as a controller to how hard the model should work to push dissimilar embedding vectors apart. For the proposed Categorical Loss, by the objective of compacting the same class labels and separating different class labels in the normalized log domain.We select penalty for those predictions with high divergence from ground-truth labels.So, we shorten correct feature vectors and enlarge false prediction tensors, it means we assign more weights for those classes with conjunction to each other (namely, “hard labels to learn”). By doing this work, we constrain the model to generate more discriminate feature vectors for variant class labels. Finally, for the proposed optimizer, our focus is on solving weak convergence of Adam optimizer for a non-convex problem. Our noteworthy optimizer is working by an alternative updating gradient procedure with an exponential weighted moving average function for faster convergence and exploiting a weight decay method to help drastically reducing the learning rate near optima to reach the dominant local minimum. We demonstrate the superiority of our proposed work by surpassing the first rank of three widely used Facial Expression Recognition datasets with 93.30% on FER-2013, and 16% improvement compare to the first rank after 10 years, reaching to 90.73% on RAF-DB, and 100% k-fold average accuracy for CK+ dataset, and shown to provide a top performance to that provided by other networks, which require much larger training datasets.

Keywords: computer vision, facial expression recognition, machine learning, algorithms, depp learning, neural networks

Procedia PDF Downloads 47

Authors: Hyeongbok Kim, Lingling Zhao, Xiaohong Su

Abstract:

Deep reinforcement learning has been applied to address various problems in robotics, such as autonomous driving and unmanned aerial vehicle. However, because of the sparse reward penalty for a collision with obstacles during the navigation mission, the agent fails to learn the optimal policy or requires a long time for convergence. Therefore, using obstacles and enemy agents, in this paper, we present a curriculum-based boost learning method to effectively train compound skills during multi-agent reinforcement learning. First, to enable the agents to solve challenging tasks, we gradually increased learning difficulties by adjusting reward shaping instead of constructing different learning environments. Then, in a benchmark environment with static obstacles and moving enemy agents, the experimental results showed that the proposed curriculum learning strategy enhanced cooperative navigation and compound collision avoidance skills in uncertain environments while improving learning efficiency.

Keywords: curriculum learning, hard exploration, multi-agent reinforcement learning, robotic navigation, sparse reward

Procedia PDF Downloads 59