The management of the design process is a challenging mission; and most researchers would argue that design is linked to intentional action and it cannot emerge out of complexity. In fact, the interactions between processes, operators, and activities define an unexpected emergent behavior, which is based on complex assumptions such as non-linearity, dynamic and adaptive firm behavior. Therefore, we need a complex thinking. This article proposes to explore how we may deepen our understanding of design process as a complex adaptive system. In fact, this new understanding creates a quite challenge for researches to develop appropriate tools to support design reasoning and decision-making. In this respect, the aim of this paper is first to define the complexity of design process as a complexity of system, by matching its characteristics with those of complex adaptive systems (CAS). Second, the paper provides an agent organizational modelization of the design process in order to support its complexity by following the ASPECS methodology which is an agent-oriented software process for engineering complex systems as well as the knowledge identification of the design process using the RIOCK meta-model.
{"title":"An agent organizational method for modeling the complexity of the design process","authors":"Abla Chaouni Benabdellah, Asmaa Benghabrit, Imane Bouhaddou, Kamar Zekhnini","doi":"10.1145/3454127.3456595","DOIUrl":"https://doi.org/10.1145/3454127.3456595","url":null,"abstract":"The management of the design process is a challenging mission; and most researchers would argue that design is linked to intentional action and it cannot emerge out of complexity. In fact, the interactions between processes, operators, and activities define an unexpected emergent behavior, which is based on complex assumptions such as non-linearity, dynamic and adaptive firm behavior. Therefore, we need a complex thinking. This article proposes to explore how we may deepen our understanding of design process as a complex adaptive system. In fact, this new understanding creates a quite challenge for researches to develop appropriate tools to support design reasoning and decision-making. In this respect, the aim of this paper is first to define the complexity of design process as a complexity of system, by matching its characteristics with those of complex adaptive systems (CAS). Second, the paper provides an agent organizational modelization of the design process in order to support its complexity by following the ASPECS methodology which is an agent-oriented software process for engineering complex systems as well as the knowledge identification of the design process using the RIOCK meta-model.","PeriodicalId":432206,"journal":{"name":"Proceedings of the 4th International Conference on Networking, Information Systems & Security","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129156707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we present a new method for semi-supervised domain adaptation in remote sensing scene classification. The method is based on a pre-trained Convolutional Neural Network (CNN) model for the extraction of highly discriminative features, followed by a fully connected layer with softmax activation function that is responsible for the classification task. The weights of the fully connected layer represent prototype feature vectors for each class. These weights are divide by a temperature parameter for normalization. The whole network is trained on both the labeled and unlabeled target samples. First, the whole network is trained on the labeled source and target samples using the standard cross entropy loss to predict their correct classes. At the same time, the model is trained to learn domain invariant features using another loss function based on entropy computed over the unlabeled target samples. Unlike the standard cross entropy loss, the novel entropy loss function is computed on the predicted probabilities of the model and does not need the true labels. The proposed model combines the standard cross entropy loss and the new unlabeled samples entropy loss and optimizes them jointly. However, the new entropy loss function needs to be maximized with respect to the classification layer to learn features that are domain invariant (hence removing the data shift), and at the same time, it should be minimized with respect to the CNN feature extractor to learn discriminative feature that are clustered around the class prototypes (in other words reducing intra-class variance). To accomplish this maximization and minimization processes at the same time, we use an adversarial training approach, where we alternate between the two processes. This type of approach is called minmax entropy and the new proposed method is called Domain Adaptation CNN with MinMax Entropy (DACNN-MME). The proposed method is tested on three RS scene datasets, namely UC Merced, AID, and NWPU. The preliminary experimental results demonstrate the potential of the proposed method. Its performance is already better than several state-of-the-art methods including RevGard, ADDA, Siamese-GAN, and MSCN. With more analysis and fine-tuning of the method even better results can be achieved in the future.
{"title":"Classification of Remote Sensing scenes using Semi-Supervised Domain Adaptation based on Entropy Adversarial Optimization","authors":"Tariq Lasloum, H. Alhichri, Y. Bazi","doi":"10.1145/3454127.3456610","DOIUrl":"https://doi.org/10.1145/3454127.3456610","url":null,"abstract":"In this paper, we present a new method for semi-supervised domain adaptation in remote sensing scene classification. The method is based on a pre-trained Convolutional Neural Network (CNN) model for the extraction of highly discriminative features, followed by a fully connected layer with softmax activation function that is responsible for the classification task. The weights of the fully connected layer represent prototype feature vectors for each class. These weights are divide by a temperature parameter for normalization. The whole network is trained on both the labeled and unlabeled target samples. First, the whole network is trained on the labeled source and target samples using the standard cross entropy loss to predict their correct classes. At the same time, the model is trained to learn domain invariant features using another loss function based on entropy computed over the unlabeled target samples. Unlike the standard cross entropy loss, the novel entropy loss function is computed on the predicted probabilities of the model and does not need the true labels. The proposed model combines the standard cross entropy loss and the new unlabeled samples entropy loss and optimizes them jointly. However, the new entropy loss function needs to be maximized with respect to the classification layer to learn features that are domain invariant (hence removing the data shift), and at the same time, it should be minimized with respect to the CNN feature extractor to learn discriminative feature that are clustered around the class prototypes (in other words reducing intra-class variance). To accomplish this maximization and minimization processes at the same time, we use an adversarial training approach, where we alternate between the two processes. This type of approach is called minmax entropy and the new proposed method is called Domain Adaptation CNN with MinMax Entropy (DACNN-MME). The proposed method is tested on three RS scene datasets, namely UC Merced, AID, and NWPU. The preliminary experimental results demonstrate the potential of the proposed method. Its performance is already better than several state-of-the-art methods including RevGard, ADDA, Siamese-GAN, and MSCN. With more analysis and fine-tuning of the method even better results can be achieved in the future.","PeriodicalId":432206,"journal":{"name":"Proceedings of the 4th International Conference on Networking, Information Systems & Security","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127422476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zakaria Benlalia, Karim Abouelmehdi, A. B. Hssane, Abdellah Ezzati
Cloud computing is the provision of information technology (IT) services including servers, storage, databases, network management, etc. Like any new technology, cloud computing requires many improvements and the establishment of precise standards to avoid risks. Task scheduling can be seen as the management and handling of a set of tasks from their start to the step of execution. Scheduling is a negotiation mechanism between two objects, one representing the user (or the ap-plication) and the other the resources. In cloud computing, Task scheduling is of-ten considered as a real challenge to managers. In this paper, we will present some concepts and research papers that have proposed improvements or solutions to this challenge and we will compare some tasks scheduling algorithms in the CloudSim simulator.
{"title":"A study of task scheduling algorithms in cloud computing","authors":"Zakaria Benlalia, Karim Abouelmehdi, A. B. Hssane, Abdellah Ezzati","doi":"10.1145/3454127.3457616","DOIUrl":"https://doi.org/10.1145/3454127.3457616","url":null,"abstract":"Cloud computing is the provision of information technology (IT) services including servers, storage, databases, network management, etc. Like any new technology, cloud computing requires many improvements and the establishment of precise standards to avoid risks. Task scheduling can be seen as the management and handling of a set of tasks from their start to the step of execution. Scheduling is a negotiation mechanism between two objects, one representing the user (or the ap-plication) and the other the resources. In cloud computing, Task scheduling is of-ten considered as a real challenge to managers. In this paper, we will present some concepts and research papers that have proposed improvements or solutions to this challenge and we will compare some tasks scheduling algorithms in the CloudSim simulator.","PeriodicalId":432206,"journal":{"name":"Proceedings of the 4th International Conference on Networking, Information Systems & Security","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116700521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Various technologies have been developed to better improve lifestyles, including Wireless Sensor Networks (WSNs), used in multiple areas of research and consisting of a large number of sensor nodes. Drums in the area of interest, sensors collect data and transmit it to the base station (BS). Therefore, the key indicator for the design of WSN is the lifetime of the network. In this article, we provide an optimized GAF routing protocol that has been improved to optimize power consumption. It provides a multi-target version, which can maximize the coverage of communication between nodes to improve coverage efficiency. This is one of the key issues in the deployment of the sensor network, followed by the elected leader to minimize the proportion of active sensor nodes. The end goal is to minimize energy consumption. All these objectives are taken into account in our proposed version. The experimental results prove that in terms of performance, number of dead nodes, and energy consumption, the objectives proposed by this new version of GAF are superior to the existing basic and optimized GAF-GAF research. The practice has proven that the proposed target GAF can improve network lifetime.
{"title":"Advanced GAF routing protocol using the goal attainment method in WSN","authors":"Hanane Aznaoui, Arif Ullah, S. Raghay, Layla Aziz","doi":"10.1145/3454127.3456581","DOIUrl":"https://doi.org/10.1145/3454127.3456581","url":null,"abstract":"Various technologies have been developed to better improve lifestyles, including Wireless Sensor Networks (WSNs), used in multiple areas of research and consisting of a large number of sensor nodes. Drums in the area of interest, sensors collect data and transmit it to the base station (BS). Therefore, the key indicator for the design of WSN is the lifetime of the network. In this article, we provide an optimized GAF routing protocol that has been improved to optimize power consumption. It provides a multi-target version, which can maximize the coverage of communication between nodes to improve coverage efficiency. This is one of the key issues in the deployment of the sensor network, followed by the elected leader to minimize the proportion of active sensor nodes. The end goal is to minimize energy consumption. All these objectives are taken into account in our proposed version. The experimental results prove that in terms of performance, number of dead nodes, and energy consumption, the objectives proposed by this new version of GAF are superior to the existing basic and optimized GAF-GAF research. The practice has proven that the proposed target GAF can improve network lifetime.","PeriodicalId":432206,"journal":{"name":"Proceedings of the 4th International Conference on Networking, Information Systems & Security","volume":"18 11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126565100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：