Search results for: federated averaging

Authors: X. Gumara Rigol, I. Martínez de Apellaniz Anzuola, A. Garcia Serrano, A. Franzi Cros, O. Vidal Calbet, A. Al Maruf

Abstract:

Traditional federated architectures for data warehousing work well when corporations have existing regional data warehouses and there is a need to aggregate data at a global level. Schibsted Media Group has been maturing from a decentralised organisation into a more globalised one and needed to build both some of the regional data warehouses for some brands at the same time as the global one. In this paper, we present the architectural alternatives studied and why a custom federated approach was the notable recommendation to go further with the implementation. Although the data warehouses are logically federated, the implementation uses a single database system which presented many advantages like: cost reduction and improved data access to global users allowing consumers of the data to have a common data model for detailed analysis across different geographies and a flexible layer for local specific needs in the same place.

Keywords: data integration, data warehousing, federated architecture, Online Analytical Processing (OLAP)

Procedia PDF Downloads 212

Authors: Shuqi Zhang

Abstract:

Federated learning allows users to participate in collaborative model training without sending data to third-party servers, reducing the risk of user data privacy leakage, and is widely used in smart finance and smart healthcare. However, the distributed architecture design of federation learning itself and the existence of secure aggregation protocols make it inherently vulnerable to backdoor attacks. To solve this problem, the federated learning backdoor defense framework FLEX based on group aggregation, cluster analysis, and neuron pruning is proposed, and inter-compatibility with secure aggregation protocols is achieved. The good performance of FLEX is verified by building a horizontal federated learning framework on the CIFAR-10 dataset for experiments, which achieves 98% success rate of backdoor detection and reduces the success rate of backdoor tasks to 0% ~ 10%.

Keywords: federated learning, secure aggregation, backdoor attack, cluster analysis, neuron pruning

Procedia PDF Downloads 60

Authors: Ananya Gangavarapu

Abstract:

Convolutional Neural Networks (CNN) based models are providing diagnostic capabilities on par with the medical specialists in many specialty areas. However, collecting the medical data for training purposes is very challenging because of the increased regulations around data collections and privacy concerns around personal health data. The gathering of the data becomes even more difficult if the capture devices are edge-based mobile devices (like smartphones) with feeble wireless connectivity in rural/remote areas. In this paper, I would like to highlight Federated Learning approach to mitigate data privacy and security issues.

Keywords: deep learning in healthcare, data privacy, federated learning, training in distributed environment

Procedia PDF Downloads 108

Authors: Aofan Liu, Qianqian Tan, Burra Venkata Durga Kumar

Abstract:

Data security is of the utmost importance in the healthcare area, as sensitive patient information is constantly sent around and analyzed by many different parties. The use of federated learning, which enables data to be evaluated locally on devices rather than being transferred to a central server, has emerged as a potential solution for protecting the privacy of user information. To protect against data breaches and unauthorized access, federated learning alone might not be adequate. In this context, the application of blockchain technology could provide the system extra protection. This study proposes a distributed federated learning system that is built on blockchain technology in order to enhance security in healthcare. This makes it possible for a wide variety of healthcare providers to work together on data analysis without raising concerns about the confidentiality of the data. The technical aspects of the system, including as the design and implementation of distributed learning algorithms, consensus mechanisms, and smart contracts, are also investigated as part of this process. The technique that was offered is a workable alternative that addresses concerns about the safety of healthcare while also fostering collaborative research and the interchange of data.

Keywords: data privacy, distributed system, federated learning, machine learning

Procedia PDF Downloads 75

Authors: Zhou Jianhong

Abstract:

Model averaging is a desirable approach to deal with model uncertainty, which, however, has rarely been explored for Poisson regression. In this paper, we propose a model averaging procedure based on an unbiased estimator of the expected Kullback-Leibler distance for the Poisson regression. Simulation study shows that the proposed model average estimator outperforms some other commonly used model selection and model average estimators in some situations. Our proposed methods are further applied to a real data example and the advantage of this method is demonstrated again.

Keywords: model averaging, poission regression, Kullback-Leibler distance, statistics

Procedia PDF Downloads 483

Authors: Xun Li, Haojie Wang

Abstract:

Federated learning is a kind of distributed training and centralized training mode, which is of great value in the protection of user privacy. In order to solve the problem that the model is vulnerable to backdoor attacks in federated learning, a backdoor attack detection method based on a weighted k-medoids algorithm is proposed. First of all, this paper collates the update parameters of the client to construct a vector group, then uses the principal components analysis (PCA) algorithm to extract the corresponding feature information from the vector group, and finally uses the improved k-medoids clustering algorithm to identify the normal and backdoor update parameters. In this paper, the backdoor is implanted in the federation learning model through the model replacement attack method in the simulation experiment, and the update parameters from the attacker are effectively detected and removed by the defense method proposed in this paper.

Keywords: federated learning, backdoor attack, PCA, k-medoids, backdoor defense

Procedia PDF Downloads 77

Authors: Peiming Li

Abstract:

This paper presents an innovative blockchain-based approach to enhance the reliability and efficiency of federated learning systems. By integrating a decentralized oracle consensus mechanism into the federated learning framework, we address key challenges of data and model integrity. Our approach utilizes a network of redundant oracles, functioning as independent validators within an epoch-based training system in the federated learning model. In federated learning, data is decentralized, residing on various participants' devices. This scenario often leads to concerns about data integrity and model quality. Our solution employs blockchain technology to establish a transparent and tamper-proof environment, ensuring secure data sharing and aggregation. The decentralized oracles, a concept borrowed from blockchain systems, act as unbiased validators. They assess the contributions of each participant using a Hidden Markov Model (HMM), which is crucial for evaluating the consistency of participant inputs and safeguarding against model poisoning and malicious activities. Our methodology's distinct feature is its epoch-based training. An epoch here refers to a specific training phase where data is updated and assessed for quality and relevance. The redundant oracles work in concert to validate data updates during these epochs, enhancing the system's resilience to security threats and data corruption. The effectiveness of this system was tested using the Mnist dataset, a standard in machine learning for benchmarking. Results demonstrate that our blockchain-oriented federated learning approach significantly boosts system resilience, addressing the common challenges of federated environments. This paper aims to make these advanced concepts accessible, even to those with a limited background in blockchain or federated learning. We provide a foundational understanding of how blockchain technology can revolutionize data integrity in decentralized systems and explain the role of oracles in maintaining model accuracy and reliability.

Keywords: federated learning system, block chain, decentralized oracles, hidden markov model

Procedia PDF Downloads 27

Authors: Akash Amalan, Rui Wang, Yanqi Qiao, Emmanouil Panaousis, Kaitai Liang

Abstract:

Federated learning is an emerging concept in the domain of distributed machine learning. This concept has enabled General Adversarial Networks (GANs) to benefit from the rich distributed training data while preserving privacy. However, in a non-IID setting, current federated GAN architectures are unstable, struggling to learn the distinct features, and vulnerable to mode collapse. In this paper, we propose an architecture MULTI-FLGAN to solve the problem of low-quality images, mode collapse, and instability for non-IID datasets. Our results show that MULTI-FLGAN is four times as stable and performant (i.e., high inception score) on average over 20 clients compared to baseline FLGAN.

Keywords: federated learning, generative adversarial network, inference attack, non-IID data distribution

Procedia PDF Downloads 114

Authors: Shayan Mohajer Hamidi

Abstract:

Federated learning has emerged as a promising approach for distributed model training while preserving data privacy. However, the challenges of communication overhead, limited network resources, and slow convergence hinder its widespread adoption. On the other hand, knowledge distillation has shown great potential in compressing large models into smaller ones without significant loss in performance. In this paper, we propose an innovative framework that combines federated learning and knowledge distillation to address these challenges and enhance the efficiency of distributed learning. Our approach, called Federated Knowledge Distillation (FKD), enables multiple clients in a federated learning setting to collaboratively distill knowledge from a teacher model. By leveraging the collaborative nature of federated learning, FKD aims to improve model compression while maintaining privacy. The proposed framework utilizes a coded teacher model that acts as a reference for distilling knowledge to the client models. To demonstrate the effectiveness of FKD, we conduct extensive experiments on various datasets and models. We compare FKD with baseline federated learning methods and standalone knowledge distillation techniques. The results show that FKD achieves superior model compression, faster convergence, and improved performance compared to traditional federated learning approaches. Furthermore, FKD effectively preserves privacy by ensuring that sensitive data remains on the client devices and only distilled knowledge is shared during the training process. In our experiments, we explore different knowledge transfer methods within the FKD framework, including Fine-Tuning (FT), FitNet, Correlation Congruence (CC), Similarity-Preserving (SP), and Relational Knowledge Distillation (RKD). We analyze the impact of these methods on model compression and convergence speed, shedding light on the trade-offs between size reduction and performance. Moreover, we address the challenges of communication efficiency and network resource utilization in federated learning by leveraging the knowledge distillation process. FKD reduces the amount of data transmitted across the network, minimizing communication overhead and improving resource utilization. This makes FKD particularly suitable for resource-constrained environments such as edge computing and IoT devices. The proposed FKD framework opens up new avenues for collaborative and privacy-preserving distributed learning. By combining the strengths of federated learning and knowledge distillation, it offers an efficient solution for model compression and convergence speed enhancement. Future research can explore further extensions and optimizations of FKD, as well as its applications in domains such as healthcare, finance, and smart cities, where privacy and distributed learning are of paramount importance.

Keywords: federated learning, knowledge distillation, knowledge transfer, deep learning

Procedia PDF Downloads 42

Authors: Alan Wan

Abstract:

In recent years, the body of literature on frequentist model averaging in statistics has grown significantly. Most of this work focuses on models with different mean structures but leaves out the variance consideration. In this paper, we consider a regression model with multiplicative heteroscedasticity and develop a model averaging method that combines maximum likelihood estimators of unknown parameters in both the mean and variance functions of the model. Our weight choice criterion is based on a minimisation of a plug-in estimator of the model average estimator's squared prediction risk. We prove that the new estimator possesses an asymptotic optimality property. Our investigation of finite-sample performance by simulations demonstrates that the new estimator frequently exhibits very favourable properties compared to some existing heteroscedasticity-robust model average estimators. The model averaging method hedges against the selection of very bad models and serves as a remedy to variance function misspecification, which often discourages practitioners from modeling heteroscedasticity altogether. The proposed model average estimator is applied to the analysis of two real data sets.

Keywords: heteroscedasticity-robust, model averaging, multiplicative heteroscedasticity, plug-in, squared prediction risk

Procedia PDF Downloads 333

Authors: Jinil Patel, Sarthak Patel, Sarthak Thakkar, Deepti Saraswat

Abstract:

Recently, the COVID-19 outbreak has spread across the world, leading the World Health Organization to classify it as a global pandemic. To save the patient’s life, the COVID-19 symptoms have to be identified. But using an AI (Artificial Intelligence) model to identify COVID-19 symptoms within the allotted time was challenging. The RT-PCR test was found to be inadequate in determining the COVID status of a patient. To determine if the patient has COVID-19 or not, a Computed Tomography Scan (CT scan) of patient is a better alternative. It will be challenging to compile and store all the data from various hospitals on the server, though. Federated learning, therefore, aids in resolving this problem. Certain deep learning models help to classify Covid-19. This paper will have detailed work of certain deep learning models like VGG19, ResNet50, MobileNEtv2, and Deep Learning Aggregation (DLA) along with maintaining privacy with encryption.

Keywords: federated learning, COVID-19, CT-scan, homomorphic encryption, ResNet50, VGG-19, MobileNetv2, DLA

Procedia PDF Downloads 42

Authors: Beulah Rani Inbanathan

Abstract:

In the present era, it is vivid from the numerous outcomes that data privacy is being compromised in various ways. Machine learning is one technology that uses the centralized server, and then data is given as input which is being analyzed by the algorithms present on this mentioned server, and hence outputs are predicted. However, each time the data must be sent by the user as the algorithm will analyze the input data in order to predict the output, which is prone to threats. The solution to overcome this issue is federated learning, where the models alone get updated while the data resides on the local machine and does not get exchanged with the other local models. Nevertheless, even on these local models, there are chances of data poisoning, and it is crystal clear from various experiments done by many people. This paper delves into many ways where data poisoning occurs and the many methods through which it is prevalent that data poisoning still exists. It includes the poisoning attacks on IoT devices, Edge devices, Autoregressive model, and also, on Industrial IoT systems and also, few points on how these could be evadible in order to protect our data which is personal, or sensitive, or harmful when exposed.

Keywords: data poisoning, federated learning, Internet of Things, edge computing

Procedia PDF Downloads 54

Authors: Ezil Sam Leni, Shalen S.

Abstract:

Transport service monitoring and upkeep are essential components of smart city initiatives. The main risks to the relevant departments and authorities are the ever-increasing vehicular traffic and the conditions of the roads. In India, the economy is greatly impacted by the road transport sector. In 2021, the Ministry of Road Transport and Highways Transport, Government of India, produced a report with statistical data on traffic accidents. The data included the number of fatalities, injuries, and other pertinent criteria. This study proposes a distributed infrastructure for the monitoring, detection, and reporting of potholes to the appropriate authorities. In a distributed environment, the nodes are the edge devices, and local edge servers, and global servers. The edge devices receive the initial model to be employed from the global server. The YOLOv8 model for pothole detection is used in the edge devices. The edge devices run the pothole detection model, gather the pothole images on their path, and send the updates to the nearby edge server. The local edge server selects the clients for its aggregation process, aggregates the model updates and sends the updates to the global server. The global server collects the updates from the local edge servers, performs aggregation and derives the updated model. The updated model has the information about the potholes received from the local edge servers and notifies the updates to the local edge servers and concerned authorities for monitoring and maintenance of road conditions. The entire process is implemented in FedCV distributed environment with the implementation using the client-server model and aggregation entities. After choosing the clients for its aggregation process, the local edge server gathers the model updates and transmits them to the global server. After gathering the updates from the regional edge servers, the global server aggregates them and creates the updated model. Performance indicators and the experimentation environment are assessed, discussed, and presented. Accelerometer data may be taken into consideration for improved performance in the future development of this study, in addition to the images captured from the transportation routes.

Keywords: federated Learning, pothole detection, distributed framework, federated averaging

Procedia PDF Downloads 49

Authors: Sirapop Nuannimnoi, Hojjat Baghban, Ching-Yao Huang

Abstract:

Over the past decade, with the fast development of blockchain technology since the birth of Bitcoin, there has been a massive increase in the usage of Cryptocurrencies. Cryptocurrencies are not seen as an investment opportunity due to the market’s erratic behavior and high price volatility. With the recent success of deep reinforcement learning (DRL), portfolio management can be modeled and automated. In this paper, we propose a novel DRL-based multi-agent system to automatically make proper trading decisions on multiple cryptocurrencies and gain profits in the highly volatile cryptocurrency market. We also extend this multi-agent system with horizontal federated transfer learning for better adapting to the inclusion of new cryptocurrencies in our portfolio; therefore, we can, through the concept of diversification, maximize our profits and minimize the trading risks. Experimental results through multiple simulation scenarios reveal that this proposed algorithmic trading system can offer three promising key advantages over other systems, including maximized profits, minimized risks, and adaptability.

Keywords: cryptocurrency portfolio management, algorithmic trading, federated learning, multi-agent reinforcement learning

Procedia PDF Downloads 85

Authors: A. Dinelka Panagoda, Bathiya Bandara, Chamod Wijetunga, Chathura Malinda, Lakmal Rupasinghe, Chethana Liyanapathirana

Abstract:

This research takes us forward with the concepts of Federated Learning and Software-Defined Networking (SDN) to introduce an efficient malware detection technique and provide a mitigation mechanism to give birth to a resilient and automated healthcare sector network system by also adding the feature of extended privacy preservation. Due to the daily transformation of new malware attacks on hospital Integrated Clinical Environment (ICEs), the healthcare industry is at an undefinable peak of never knowing its continuity direction. The state of blindness by the array of indispensable opportunities that new medical device inventions and their connected coordination offer daily, a factor that should be focused driven is not yet entirely understood by most healthcare operators and patients. This solution has the involvement of four clients in the form of hospital networks to build up the federated learning experimentation architectural structure with different geographical participation to reach the most reasonable accuracy rate with privacy preservation. While the logistic regression with cross-entropy conveys the detection, SDN comes in handy in the second half of the research to stack up the initial development phases of the system with malware mitigation based on policy implementation. The overall evaluation sums up with a system that proves the accuracy with the added privacy. It is no longer needed to continue with traditional centralized systems that offer almost everything but not privacy.

Keywords: software-defined network, federated learning, privacy, integrated clinical environment, decentralized learning, malware detection, malware mitigation

Procedia PDF Downloads 137

Authors: Yaping Zhao

Abstract:

In the present study, the exact solutions for the steady response of quasi-linear systems under non-white wide-band random excitation are considered by means of the stochastic averaging method. The non linearity of the systems contains the power-law damping and the cross-product term of the power-law damping and displacement. The drift and diffusion coefficients of the Fokker-Planck-Kolmogorov (FPK) equation after averaging are obtained by a succinct approach. After solving the averaged FPK equation, the joint probability density function and the marginal probability density function in steady state are attained. In the process of resolving, the eigenvalue problem of ordinary differential equation is handled by integral equation method. Some new results are acquired and the novel method to deal with the problems in nonlinear random vibration is proposed.

Keywords: random vibration, stochastic averaging method, FPK equation, transition probability density

Procedia PDF Downloads 469

Authors: Shayan Mohajer Hamidi

Abstract:

Federated Learning (FL) has emerged as a promising paradigm for training machine learning models on distributed data without the need for centralized data aggregation. In the realm of wireless communications, FL has the potential to leverage the vast amounts of data generated by wireless devices to improve model performance and enable intelligent applications. However, the fairness aspect of FL in wireless communications remains largely unexplored. This abstract presents an idea for fair federated learning in wireless communications, addressing the challenges of imbalanced data distribution, privacy preservation, and resource allocation. Firstly, the proposed approach aims to tackle the issue of imbalanced data distribution in wireless networks. In typical FL scenarios, the distribution of data across wireless devices can be highly skewed, resulting in unfair model updates. To address this, we propose a weighted aggregation strategy that assigns higher importance to devices with fewer samples during the aggregation process. By incorporating fairness-aware weighting mechanisms, the proposed approach ensures that each participating device's contribution is proportional to its data distribution, thereby mitigating the impact of data imbalance on model performance. Secondly, privacy preservation is a critical concern in federated learning, especially in wireless communications where sensitive user data is involved. The proposed approach incorporates privacy-enhancing techniques, such as differential privacy, to protect user privacy during the model training process. By adding carefully calibrated noise to the gradient updates, the proposed approach ensures that the privacy of individual devices is preserved without compromising the overall model accuracy. Moreover, the approach considers the heterogeneity of devices in terms of computational capabilities and energy constraints, allowing devices to adaptively adjust the level of privacy preservation to strike a balance between privacy and utility. Thirdly, efficient resource allocation is crucial for federated learning in wireless communications, as devices operate under limited bandwidth, energy, and computational resources. The proposed approach leverages optimization techniques to allocate resources effectively among the participating devices, considering factors such as data quality, network conditions, and device capabilities. By intelligently distributing the computational load, communication bandwidth, and energy consumption, the proposed approach minimizes resource wastage and ensures a fair and efficient FL process in wireless networks. To evaluate the performance of the proposed fair federated learning approach, extensive simulations and experiments will be conducted. The experiments will involve a diverse set of wireless devices, ranging from smartphones to Internet of Things (IoT) devices, operating in various scenarios with different data distributions and network conditions. The evaluation metrics will include model accuracy, fairness measures, privacy preservation, and resource utilization. The expected outcomes of this research include improved model performance, fair allocation of resources, enhanced privacy preservation, and a better understanding of the challenges and solutions for fair federated learning in wireless communications. The proposed approach has the potential to revolutionize wireless communication systems by enabling intelligent applications while addressing fairness concerns and preserving user privacy.

Keywords: federated learning, wireless communications, fairness, imbalanced data, privacy preservation, resource allocation, differential privacy, optimization

Procedia PDF Downloads 42

Authors: Xin Li, Zhenxue Jiang, Zhuo Li

Abstract:

Generally, shale gas is trapped within shale systems with low porosity and ultralow permeability as free and adsorbing states. Its production is controlled by properties, in terms of occurrence phases, gas contents, and percolation characteristics. These properties are all influenced by porous features. In this paper, porosity differences of marine shales were explored between Lower Cambrian shale and Lower Silurian shale of Sichuan Basin, South China. Both the two shales were marine shales with abundant oil-prone kerogen and rich siliceous minerals. Whereas Lower Cambrian shale (3.56% Ro) possessed a higher thermal degree than that of Lower Silurian shale (2.31% Ro). Samples were measured by a combination of organic-chemistry geology measurement, organic matter (OM) isolation, X-ray diffraction (XRD), N2 adsorption, and focused ion beam milling and scanning electron microscopy (FIB-SEM). Lower Cambrian shale presented relatively low pore properties, with averaging 0.008ml/g pore volume (PV), averaging 7.99m²/g pore surface area (PSA) and averaging 5.94nm average pore diameter (APD). Lower Silurian shale showed as relatively high pore properties, with averaging 0.015ml/g PV, averaging 10.53m²/g PSA and averaging 18.60nm APD. Additionally, fractal analysis indicated that the two shales presented discrepant pore morphologies, mainly caused by differences in the combination of pore types between the two shales. More specifically, OM-hosted pores with pin-hole shape and dissolved pores with dead-end openings were the main types in Lower Cambrian shale, while OM-hosted pore with a cellular structure was the main type in Lower Silurian shale. Moreover, porous characteristics of isolated OM suggested that OM of Lower Silurian shale was more capable than that of Lower Cambrian shale in the aspect of pore contribution. PV of isolated OM in Lower Silurian shale was almost 6.6 times higher than that in Lower Cambrian shale, and PSA of isolated OM in Lower Silurian shale was almost 4.3 times higher than that in Lower Cambrian shale. However, no apparent differences existed among samples with various matrix compositions. At late diagenetic or metamorphic epoch, extensive diagenesis overprints the effects of minerals on pore properties and OM plays the dominant role in pore developments. Hence, differences of porous features between the two marine shales highlight the effect of diagenetic degree on OM-hosted pore development. Consequently, distinctive pore characteristics may be caused by the different degrees of diagenetic evolution, even with similar matrix basics.

Keywords: marine shale, lower Cambrian, lower Silurian, om isolation, pore properties, om-hosted pore

Procedia PDF Downloads 108

Authors: Violeta Damjanovic-Behrendt

Abstract:

This paper presents an approach for optimal cyber security decisions to protect instances of a federated Internet of Things (IoT) platform in the cloud. The presented solution implements the repeated Stackelberg Security Game (SSG) and a model called Stochastic Human behaviour model with AttRactiveness and Probability weighting (SHARP). SHARP employs the Subjective Utility Quantal Response (SUQR) for formulating a subjective utility function, which is based on the evaluations of alternative solutions during decision-making. We augment the repeated SSG (including SHARP and SUQR) with a reinforced learning algorithm called Naïve Q-Learning. Naïve Q-Learning belongs to the category of active and model-free Machine Learning (ML) techniques in which the agent (either the defender or the attacker) attempts to find an optimal security solution. In this way, we combine GT and ML algorithms for discovering optimal cyber security policies. The proposed security optimization components will be validated in a collaborative cloud platform that is based on the Industrial Internet Reference Architecture (IIRA) and its recently published security model.

Keywords: security, internet of things, cloud computing, stackelberg game, machine learning, naive q-learning

Procedia PDF Downloads 324

Authors: Muhammad Luthfi, Sutikno Sutikno, Purhadi Purhadi

Abstract:

Weather forecast has necessarily been improved to provide the communities an accurate and objective prediction as well. To overcome such issue, the numerical-based weather forecast was extensively developed to reduce the subjectivity of forecast. Yet the Numerical Weather Predictions (NWPs) outputs are unfortunately issued without taking dynamical weather behavior and local terrain features into account. Thus, NWPs outputs are not able to accurately forecast the weather quantities, particularly for medium and long range forecast. The aim of this research is to aid and extend the development of ensemble forecast for Meteorology, Climatology, and Geophysics Agency of Indonesia. Ensemble method is an approach combining various deterministic forecast to produce more reliable one. However, such forecast is biased and uncalibrated due to its underdispersive or overdispersive nature. As one of the parametric methods, Bayesian Model Averaging (BMA) generates the calibrated ensemble forecast and constructs predictive PDF for specified period. Such method is able to utilize ensemble of any size but does not take spatial correlation into account. Whereas space dependencies involve the site of interest and nearby site, influenced by dynamic weather behavior. Meanwhile, Geostatistical Output Perturbation (GOP) reckons the spatial correlation to generate future weather quantities, though merely built by a single deterministic forecast, and is able to generate an ensemble of any size as well. This research conducts both BMA and GOP to generate the calibrated ensemble forecast for the daily temperature at few meteorological sites nearby Indonesia international airport.

Keywords: Bayesian Model Averaging, ensemble forecast, geostatistical output perturbation, numerical weather prediction, temperature

Procedia PDF Downloads 246

Authors: Yoonsuh Jung

Abstract:

As DNA microarray data contain relatively small sample size compared to the number of genes, high dimensional models are often employed. In high dimensional models, the selection of tuning parameter (or, penalty parameter) is often one of the crucial parts of the modeling. Cross-validation is one of the most common methods for the tuning parameter selection, which selects a parameter value with the smallest cross-validated score. However, selecting a single value as an "optimal" value for the parameter can be very unstable due to the sampling variation since the sample sizes of microarray data are often small. Our approach is to choose multiple candidates of tuning parameter first, then average the candidates with different weights depending on their performance. The additional step of estimating the weights and averaging the candidates rarely increase the computational cost, while it can considerably improve the traditional cross-validation. We show that the selected value from the suggested methods often lead to stable parameter selection as well as improved detection of significant genetic variables compared to the tradition cross-validation via real data and simulated data sets.

Keywords: cross validation, parameter averaging, parameter selection, regularization parameter search

Procedia PDF Downloads 383

Authors: Shaho Ghafur Ahmed

Abstract:

The right to self-determination of peoples is a fundamental human right recognized by the principles of international law. It could be embodied in the internal level in the form of federalism. Most federal constitutions prevented the secession of constituent entities, while some remained silent, as the case of Iraq, and rare instances of them recognize it. But, after the failure of federalism, these entities seek to separate whenever the opportunity arises. In several cases, they have resort to peaceful methods in some others they resort to force. The constitutional Supreme Court, which guaranty the unity and integrity of the State, often prevent these attempts. After not a commitment of federalism in Iraq, which has been founded since 2004, the Kurdistan region, as the only federated entity, has conducted a unilateral referendum on 25 September 2017 for its independence. The Iraqi government refused it. The Iraqi Federal Supreme Court, through interpreting the constitutional provisions, decided that this referendum and it’s purposes, which was the independence of the region, was unconstitutional. Subsequently, the Iraqi government used forces and blockaded the region so as to force it to turn off this process. So, in this paper, the right to self-determination of the peoples in federated entities and its obstacles will be discussed through the comparative legal basis and analyzing the decisions of the Federal Constitutional Courts. We will compare the role that the Supreme Court of Canada played regarding the referendum that operated in Quebec in 1995, in which it refused only the unilaterally attempts for the independence of this province. While, in the case of the Kurdistan region, the Iraqi Federal Supreme Court has definitively refused this right. No measures were taken by this Court to protect the region from the Iraqi government reactions. This decision led to the questioning of the neutrality of this Court. So, from the point of view of the Kurdistan region, this Court became a political instrument to prevent it to be independent in the international community, in the absence of a clear constitutional provision, through an abstract and an incomplete interpretation of federal constitutional provisions.

Keywords: right of self-determination, federal supreme court, supremacy of federal constitution

Procedia PDF Downloads 136

Authors: Jesus Carretero, Mario Vasile, Guillermo Izquierdo, Javier Garcia-Blas

Abstract:

Currently, there is a so called ‘identity crisis’ in cybersecurity caused by the substantial security, privacy and usability shortcomings encountered in existing systems for identity management. Federated Identity Management (FIM) could be solution for this crisis, as it is a method that facilitates management of identity processes and policies among collaborating entities without enforcing a global consistency, that is difficult to achieve when there are ID legacy systems. To cope with this problem, the Connecting Europe Facility (CEF) initiative proposed in 2014 a federated solution in anticipation of the adoption of the Regulation (EU) N°910/2014, the so-called eIDAS Regulation. At present, a network of eIDAS Nodes is being deployed at European level to allow that every citizen recognized by a member state is to be recognized within the trust network at European level, enabling the consumption of services in other member states that, until now were not allowed, or whose concession was tedious. This is a very ambitious approach, since it tends to enable cross-border authentication of Member States citizens without the need to unify the authentication method (eID Scheme) of the member state in question. However, this federation is currently managed by member states and it is initially applied only to citizens and public organizations. The goal of this paper is to present the results of a European Project, named eID@Cloud, that focuses on the integration of eID in 5 cloud platforms belonging to authentication service providers of different EU Member States to act as Service Providers (SP) for private entities. We propose an initiative based on a private eID Scheme both for natural and legal persons. The methodology followed in the eID@Cloud project is that each Identity Provider (IdP) is subscribed to an eIDAS Node Connector, requesting for authentication, that is subscribed to an eIDAS Node Proxy Service, issuing authentication assertions. To cope with high loads, load balancing is supported in the eIDAS Node. The eID@Cloud project is still going on, but we already have some important outcomes. First, we have deployed the federation identity nodes and tested it from the security and performance point of view. The pilot prototype has shown the feasibility of deploying this kind of systems, ensuring good performance due to the replication of the eIDAS nodes and the load balance mechanism. Second, our solution avoids the propagation of identity data out of the native domain of the user or entity being identified, which avoids problems well known in cybersecurity due to network interception, man in the middle attack, etc. Last, but not least, this system allows to connect any country or collectivity easily, providing incremental development of the network and avoiding difficult political negotiations to agree on a single authentication format (which would be a major stopper).

Keywords: cybersecurity, identity federation, trust, user authentication

Procedia PDF Downloads 139

Authors: Lingerew Nebere Kassie, Ping-Yu Chang, Hsin-Hua Huang, , Chaw-Son Chen

Abstract:

Transient electromagnetic (TEM) data plays a crucial role in hydrogeological and environmental applications, providing valuable insights into geological structures and resistivity variations. However, the presence of noise often hinders the interpretation and reliability of these data. Our study addresses this issue by utilizing a FASTSNAP system for the TEM survey, which operates at different modes (low, medium, and high) with continuous adjustments to discretization, gain, and current. We employ a denoising approach that processes the raw data obtained from each acquisition mode to improve signal quality and enhance data reliability. We use a signal-averaging technique for each mode, increasing the signal-to-noise ratio. Additionally, we utilize wavelet transform to suppress noise further while preserving the integrity of the underlying signals. This approach significantly improves the data quality, notably suppressing severe noise at late times. The resulting denoised data exhibits a substantially improved signal-to-noise ratio, leading to increased accuracy in parameter estimation. By effectively denoising TEM data, our study contributes to a more reliable interpretation and analysis of underground structures. Moreover, the proposed denoising approach can be seamlessly integrated into existing ground-based TEM data processing workflows, facilitating the extraction of meaningful information from noisy measurements and enhancing the overall quality and reliability of the acquired data.

Keywords: data quality, signal averaging, transient electromagnetic, wavelet transform

Procedia PDF Downloads 54

Authors: Adandé Belarmain Fandohan

Abstract:

Climate change has become a major concern globally, particularly in rural communities that have to find rapid coping solutions. Several vulnerability assessment approaches have been developed in the last decades. This comes along with a higher risk for different methods to result in different conclusions, thereby making comparisons difficult and decision-making non-consistent across areas. The effect of methods and computational approaches on estimates of people’s vulnerability was assessed using data collected from the Gambia. Twenty-four indicators reflecting vulnerability components: (exposure, sensitivity, and adaptive capacity) were selected for this purpose. Data were collected through household surveys and key informant interviews. One hundred and fifteen respondents were surveyed across six communities and two administrative districts. Results were compared over three computational approaches: the maximum value transformation normalization, the z-score transformation normalization, and simple averaging. Regardless of the approaches used, communities that have high exposure to climate change and extreme events were the most vulnerable. Furthermore, the vulnerability was strongly related to the socio-economic characteristics of farmers. The survey evidenced variability in vulnerability among communities and administrative districts. Comparing output across approaches, overall, people in the study area were found to be highly vulnerable using the simple average and maximum value transformation, whereas they were only moderately vulnerable using the z-score transformation approach. It is suggested that assessment approach-induced discrepancies be accounted for in international debates to harmonize/standardize assessment approaches to the end of making outputs comparable across regions. This will also likely increase the relevance of decision-making for adaptation policies.

Keywords: maximum value transformation, simple averaging, vulnerability assessment, West Africa, z-score transformation

Procedia PDF Downloads 75

Authors: A. Elmethnani, A. Jroud

Abstract:

This coastal crude oil refinery is situated north of a big city west of Tripoli; the city then could be highly prone to downwind refinery emissions where the NNE wind direction is prevailing through most seasons of the year. Furthermore, due to the absence of an air quality monitoring network and scarce emission data available for the neighboring community, nearby residents have serious worries about the impacts of the oil refining operations on local air quality. In responding to these concerns, a short term survey has performed for three consecutive days where a semi-continues mobile monitoring approach has developed effectively in this study; the monitoring station (Compact AQM 65 AeroQual) was mounted on a vehicle to move quickly between locations, measurements of 10 minutes averaging of 60 seconds then been taken at each fixed sampling point. The downwind ambient concentration of CO, H₂S, NOₓ, NO₂, SO₂, PM₁, PM₂.₅ PM₁₀, and TSP were measured at carefully chosen sampling locations, ranging from 200m nearby the fence-line passing through the city center up to 4.7 km east to attain best spatial coverage. Results showed worrying levels of PM₂.₅ PM₁₀, and TSP at one sampling location in the city center, southeast of the refinery site, with an average mean of 16.395μg/m³, 33.021μg/m³, and 42.426μg/m³ respectively, which could be attributed to road traffic. No significant concentrations have been detected for other pollutants of interest over the study area, as levels observed for CO, SO₂, H₂S, NOₓ, and NO₂ haven’t respectively exceeded 1.707 ppm, 0.021ppm, 0.134 ppm, 0.4582 ppm, and 0.0018 ppm, which was at the same sampling locations as well. Although it wasn’t possible to compare the results with the Libyan air quality standards due to the difference in the averaging time period, the technique was adequate for the baseline air quality screening procedure. Overall, findings primarily suggest modeling of dispersion of the refinery emissions to assess the likely impact and spatial-temporal distribution of air pollutants.

Keywords: air quality, mobil monitoring, oil refinery

Procedia PDF Downloads 67

Authors: Moatlhodi Wise Letshwenyo, Lesedi Lebogang

Abstract:

Botswana is an arid country that needs to start reusing wastewater as part of its water security plan. Pilot scale slow sand filtration in combination with roughing filter was investigated for the treatment of effluent from Botswana International University of Science and Technology to meet Botswana irrigation standards. The system was operated at hydraulic loading rates of 0.04 m/hr and 0.12 m/hr. The results show that the system was able to reduce turbidity from 262 Nephelometric Turbidity Units to a range between 18 and 0 Nephelometric Turbidity Units which was below 30 Nephelometric Turbidity Units threshold limit. The overall efficacy ranged between 61% and 100%. Suspended solids, Biochemical Oxygen Demand, and Chemical Oxygen Demand removal efficiency averaged 42.6%, 45.5%, and 77% respectively and all within irrigation standards. Other physio-chemical parameters were within irrigation standards except for bicarbonate ion which averaged 297.7±44 mg L^-1 in the influent and 196.22±50 mg L^-1 in the effluent which was above the limit of 92 mg L^-1, therefore averaging a reduction of 34.1% by the system. Total coliforms, fecal coliforms, and Escherichia coli in the effluent were initially averaging 1.1 log counts, 0.5 log counts, and 1.3 log counts respectively compared to corresponding influent log counts of 3.4, 2.7 and 4.1, respectively. As time passed, it was observed that only roughing filter was able to reach reductions of 97.5%, 86% and 100% respectively for faecal coliforms, Escherichia coli, and total coliforms. These organism numbers were observed to have increased in slow sand filter effluent suggesting multiplication in the tank. Water quality index value of 22.79 for the physio-chemical parameters suggests that the effluent is of excellent quality and can be used for irrigation purposes. However, the water quality index value for the microbial parameters (1820) renders the quality unsuitable for irrigation. It is concluded that slow sand filtration in combination with roughing filter is a viable option for the treatment of secondary effluent for reuse purposes. However, further studies should be conducted especially for the removal of microbial parameters using the system.

Keywords: irrigation, slow sand filter, turbidity, wastewater reuse

Procedia PDF Downloads 124

Authors: Eva L. Suarez, Daniel E. Meeroff, Yan Yong

Abstract:

Current flooding risk modeling focuses on resilience, defined as the probability of recovery from a severe flooding event. However, the long-term damage to property and well-being by nuisance flooding and its long-term effects on communities are not typically included in risk assessments. An approach was developed to address the probability of recovering from a severe flooding event combined with the probability of community performance during a nuisance event. A consolidated model, namely the conflation flooding recovery (&FR) model, evaluates risk-coping mitigation strategies for communities based on the recovery time from catastrophic events, such as hurricanes or extreme surges, and from everyday nuisance flooding events. The &FR model assesses the variation contribution of each independent input and generates a weighted output that favors the distribution with minimum variation. This approach is especially useful if the input distributions have dissimilar variances. The &FR is defined as a single distribution resulting from the product of the individual probability density functions. The resulting conflated distribution resides between the parent distributions, and it infers the recovery time required by a community to return to basic functions, such as power, utilities, transportation, and civil order, after a flooding event. The &FR model is more accurate than averaging individual observations before calculating the mean and variance or averaging the probabilities evaluated at the input values, which assigns the same weighted variation to each input distribution. The main disadvantage of these traditional methods is that the resulting measure of central tendency is exactly equal to the average of the input distribution’s means without the additional information provided by each individual distribution variance. When dealing with exponential distributions, such as resilience from severe flooding events and from nuisance flooding events, conflation results are equivalent to the weighted least squares method or best linear unbiased estimation. The combination of severe flooding risk with nuisance flooding improves flood risk management for highly populated coastal communities, such as in South Florida, USA, and provides a method to estimate community flood recovery time more accurately from two different sources, severe flooding events and nuisance flooding events.

Keywords: community resilience, conflation, flood risk, nuisance flooding

Procedia PDF Downloads 58

Authors: Anuja Askhedkar, G. H. Agrawal, Madhu Gudgunti

Abstract:

Direct Charge Transfer Switched Capacitor (DCT-SC) DAC is the internal DAC used in Delta-Sigma (∆∑) DAC which works on Over-Sampling concept. The Switched Capacitor DAC mainly suffers from mismatch among capacitors. Mismatch among capacitors in DAC, causes non linearity between output and input. Dynamic Element Matching (DEM) technique is used to match the capacitors. According to element selection logic there are many types. In this paper, Data Weighted Averaging (DWA) technique is used for mismatch shaping. In this paper, the 4 bit DCT-SC-DAC with DWA-DEM technique is implemented using WINSPICE simulation software in 180nm CMOS technology. DNL for DAC with DWA is ±0.03 LSB and INL is ± 0.02LSB.

Keywords: ∑-Δ DAC, DCT-SC-DAC, mismatch shaping, DWA, DEM

Procedia PDF Downloads 322

Authors: Lev Chernyshev, Ekaterina Lieshout, Natalia Kabaliuk

Abstract:

Sit-skis enable individuals with limited lower limb or core movement to ski unassisted confidently. The rise in popularity of the Winter Paralympics has seen an influx of engineering innovation, especially for the Downhill and Super-Giant Slalom events, where the athletes achieve speeds as high as 160km/h. The growth in the sport has inspired recent research into sit-ski aerodynamics. Crosswinds are expected in mountain climates and, therefore, can greatly impact a skier's maneuverability and aerodynamics. This research investigates the impact of crosswinds on the drag force of a Paralympic sit-ski using Computational Fluid Dynamics (CFD). A Paralympic sit-ski with a model of a skier, a leg cover, a bucket seat, and a simplified suspension system was used for CFD analysis in ANSYS Fluent. The hybrid initialisation tool and the SST k–ω turbulence model were used with two tetrahedral mesh bodies of influence. The crosswinds (10, 30, and 50 km/h) acting perpendicular to the sit-ski's direction of travel were simulated, corresponding to the straight-line skiing speeds of 60, 80, and 100km/h. Following the initialisation, 150 iterations for both first and second order steady-state solvers were used, before switching to a transient solver with a computational time of 1.5s and a time step of 0.02s, to allow the solution to converge. CFD results were validated against wind tunnel data. The results suggested that for all crosswind and sit-ski speeds, on average, 64% of the total drag on the ski was due to the athlete's torso. The suspension was associated with the second largest overall sit-ski drag force contribution, averaging at 27%, followed by the leg cover at 10%. While the seat contributed a negligible 0.5% of the total drag force, averaging at 1.2N across the conditions studied. The effect of the crosswind increased the total drag force across all skiing speed studies, with the drag on the athlete's torso and suspension being the most sensitive to the changes in the crosswind magnitude. The effect of the crosswind on the ski drag reduced as the simulated skiing speed increased: for skiing at 60km/h, the drag force on the torso increased by 154% with the increase of the crosswind from 10km/h to 50km/h; whereas, at 100km/h the corresponding drag force increase was halved (75%). The analysis of the flow and pressure field characteristics for a sit-ski in crosswind conditions indicated the flow separation localisation and wake size correlated with the magnitude and directionality of the crosswind relative to straight-line skiing. The findings can inform aerodynamic improvements in sit-ski design and increase skiers' medalling chances.

Keywords: sit-ski, aerodynamics, CFD, crosswind effects

Procedia PDF Downloads 44