Imitation learning is an effective method for training an autonomous agent to accomplish a task by imitating expert behaviors in their demonstrations. However, traditional imitation learning methods require a large number of expert demonstrations in order to learn a complex behavior. Such a disadvantage has limited the potential of imitation learning in complex tasks where the expert demonstrations are not sufficient. In order to address the problem, we propose a Generative Adversarial Network-based model which is designed to learn optimal policies using only a single demonstration. The proposed model is evaluated on two simulated tasks in comparison with other methods. The results show that our proposed model is capable of completing co

The research focuses on determination of best location of high elevated tank using the required head of pump as a measure for this purpose. Five types of network were used to find the effect of the variation in the discharge and the node elevation on the best location. The most weakness point was determined for each network. Preliminary tank locations were chosen for test along the primary pipe with same interval distance. For each location, the water elevation in tank and pump head was calculated at each hour depending on the pump head that required to achieve the minimum pressure at the most weakness point. Then, the sum of pump heads through the day was determined. The results proved that there is a most economical lo

In the last two decades, networks had been changed according to the rapid changing in its requirements. The current Data Center Networks have large number of hosts (tens or thousands) with special needs of bandwidth as the cloud network and the multimedia content computing is increased. The conventional Data Center Networks (DCNs) are highlighted by the increased number of users and bandwidth requirements which in turn have many implementation limitations. The current networking devices with its control and forwarding planes coupling result in network architectures are not suitable for dynamic computing and storage needs. Software Defined networking (SDN) is introduced to change this notion of traditional networks by decoupling control and

The Internet is providing vital communications between millions of individuals. It is also more and more utilized as one of the commerce tools; thus, security is of high importance for securing communications and protecting vital information. Cryptography algorithms are essential in the field of security. Brute force attacks are the major Data Encryption Standard attacks. This is the main reason that warranted the need to use the improved structure of the Data Encryption Standard algorithm. This paper proposes a new, improved structure for Data Encryption Standard to make it secure and immune to attacks. The improved structure of Data Encryption Standard was accomplished using standard Data Encryption Standard with a new way of two key gene

Image classification is the process of finding common features in images from various classes and applying them to categorize and label them. The main problem of the image classification process is the abundance of images, the high complexity of the data, and the shortage of labeled data, presenting the key obstacles in image classification. The cornerstone of image classification is evaluating the convolutional features retrieved from deep learning models and training them with machine learning classifiers. This study proposes a new approach of “hybrid learning” by combining deep learning with machine learning for image classification based on convolutional feature extraction using the VGG-16 deep learning model and seven class

A substantial portion of today’s multimedia data exists in the form of unstructured text. However, the unstructured nature of text poses a significant task in meeting users’ information requirements. Text classification (TC) has been extensively employed in text mining to facilitate multimedia data processing. However, accurately categorizing texts becomes challenging due to the increasing presence of non-informative features within the corpus. Several reviews on TC, encompassing various feature selection (FS) approaches to eliminate non-informative features, have been previously published. However, these reviews do not adequately cover the recently explored approaches to TC problem-solving utilizing FS, such as optimization techniques.

The aesthetic contents of data visualization is one of the contemporary areas through which data scientists and designers have been able to link data to humans, and even after reaching successful attempts to model data visualization, it wasn't clear how that reveals how it contributed to choosing the aesthetic content as an input to humanize these models, so the goal of the current research is to use The analytical descriptive approach aims to identify the aesthetic contents in data visualization, which the researchers interpreted through pragmatic philosophy and Kantian philosophy, and analyze a sample of data visualization models to reveal the aesthetic entrances in them to explain how to humanize them. The two researchers reached seve

With wireless sensor network (WSN) wide applications in popularity, securing its data becomes a requirement. This can be accomplished by encrypting sensor node data. In this paper a new an efficient symmetric cryptographic algorithm is presented. This algorithm is called wireless sensor network wavelet curve ciphering system (WSN-WCCS).  The algorithm idea based on discrete wavelet transformation to generate keys for each node in WSN.  It implements on hierarchical clustering WSN using LEACH protocol. Python programming language version 2.7 was used to create the simulator of WSN framework and implement a WSN-WCCS algorithm. The simulation result of the proposed WSN-WCCS with other symmetric algorithms has show

Multiple eliminations (de-multiple) are one of seismic processing steps to remove their effects and delineate the correct primary refractors. Using normal move out to flatten primaries is the way to eliminate multiples through transforming these data to frequency-wavenumber domain. The flatten primaries are aligned with zero axis of the frequency-wavenumber domain and any other reflection types (multiples and random noise) are distributed elsewhere. Dip-filter is applied to pass the aligned data and reject others will separate primaries from multiple after transforming the data back from frequency-wavenumber domain to time-distance domain. For that, a suggested name for this technique as normal move out- frequency-wavenumber domain