Search

Learning Automata (LAs) are adaptive decision-making models designed to find an appropriate action in unknown environments. LAs can be classified into two classes: variable structure and fixed structure. To the best of our knowledge, there is no hybrid model based on both of these classes. In this paper, we propose a model that brings together the benefits of both classes of LAs. In the proposed model, called an HLA, the action switching phase of a fixed structure learning automaton is fused with a variable structure learning automaton. Several computer simulations are conducted to study the performance of the proposed model with respect to the total number of rewards and action switching in addition to the convergence rate. The proposed model is compared to both variable structure and fixed structure learning automata, and in most cases, the numerical results demonstrate its superiority. In order to show the applicability of the HLA, a novel adaptive dropout mechanism in deep neural networks was suggested. The results of the simulations show that the proposed mechanism performs better than the simple dropout mechanism with respect to network accuracy. [ABSTRACT FROM AUTHOR]

Summary: Cognitive peer‐to‐peer networks are obtained from a combination of cognitive networking concepts and peer‐to‐peer networks. These networks are able to improve their performance while operating under dynamic and unknown environments. A cognitive peer‐to‐peer network tries to learn an appropriate configuration for itself considering the unknown physical properties of peers. Cognitive mobile peer‐to‐peer networks refer to cognitive peer‐to‐peer networks which are built over mobile ad hoc networks. In these networks, heterogeneity of the mobility of peers and resource limitation in wireless networks create challenges for network management algorithms. Because of the dynamicity of these networks, the management algorithms should be designated in self‐adaptive manner. In one type of these networks, some peers, called super‐peers, undertake to perform network managerial tasks. The mobility of peers leads to connection failure among peers and reselection of new super‐peers. Therefore, the selection of super‐peers, due to their influential role, requires an algorithm that considers the peers' mobility. Up to now, no self‐adaptive algorithm has been designated for super‐peer selection considering the mobility of peers in a self‐adaptive manner. This paper proposes M‐SSBLA, a self‐adaptive algorithm for super‐peer selection considering the mobility of peers based on learning automata. The proposed algorithm is obtained from cooperation between a learning automata‐based cognitive engine and MIS. MIS is a well‐known super‐peer selection algorithm in mobile peer‐to‐peer networks. We compared the proposed algorithm with recently reported algorithms, especially for a network with high mobility. Simulation results show that the proposed algorithm can cover maximum ordinary‐peer with a few super‐peer and improve robustness against super‐peer failures while decreasing maintenance overhead. [ABSTRACT FROM AUTHOR]

Cellular learning automata (CLAs) are learning models that bring together the computational power of cellular automata and also the learning capability of learning automata in unknown environments. CLAs can be open or closed. In a closed CLA , the action of each learning automaton depends on the neighboring cells, whereas in an open CLA , the action of each learning automaton depends on the neighboring cells, and a global environment. These models can be synchronous or asynchronous . In a synchronous CLA , all cells are activated at the same time, but in an asynchronous CLA , at a given time only some cells are activated. These models can be also static or dynamic . In a dynamic CLA, one of its aspects such as structure, local rule or neighborhood may vary with time . All existing dynamic models of the CLAs are closed. In this paper, an open asynchronous dynamic CLA has been introduced. In order to show the potential of this model, an algorithm based on this model for solving allocation hub location problem with imprecise distances among nodes has been designed. To evaluate the proposed algorithm computer simulations have been conducted. The results of simulations show that the proposed algorithm is more robust to imprecise distances as compared to existing algorithms. [ABSTRACT FROM AUTHOR]

Mobile peer-to-peer (MP2P) networks refer to the peer-to-peer overlay networks superimposing above the mobile ad-hoc networks. Heterogeneity of capacity and mobility of the peers as well as inherent limitation of resources along with the wireless networks characteristics are challenges on MP2P networks. In some MP2P networks, in order to improve network performances, special peers, are called super-peers, undertake to perform network managerial tasks. Selection of super-peers, due to their influential position, requires a protocol which considers the capacity of peers. Lack of general information about the capacity of other peers, as well as peers mobility along with dynamic nature of MP2P networks are the major challenges that impose uncertainty in decision making of the super-peer management algorithms. This paper proposes an adaptive super-peer selection algorithm considering peers capacity based on learning automata in MP2P networks, called SSBLA. In the proposed algorithm, each peer is equipped with a learning automaton which is used locally in the operation of super-peer selection by that peer. It has been shown that the suggested algorithm is superior to the existing algorithms. The results of the simulation show that the proposed algorithm can maximize capacity utilization by minimum number of super-peer and improve robustness against failures of super-peers while minimizing selection communication overhead. [ABSTRACT FROM AUTHOR]

Peer-to-peer networks are overlay networks that are built on top of communication networks that are called underlay networks. In these networks, peers are unaware of the underlying networks, so the peers choose their neighbors without considering the underlay positions, and therefore, the resultant overlay network may have mismatches with its underlying network, causing redundant end-to-end delay. Landmark clustering algorithms, such as mOverlay, are used to solve topology mismatch problem. In the mOverlay algorithm, the overlay network is formed by clusters in which each cluster has a landmark peer. One of the drawbacks of mOverlay is that the selected landmark peer for each cluster is fixed during the operation of the network. Because of the dynamic nature of peer-to-peer networks, using a non-adaptive landmark selection algorithm may not be appropriate. In this paper, an adaptive landmark clustering algorithm obtained from the combination of mOverlay and learning automata is proposed. Learning automata are used to adaptively select appropriate landmark peers for the clusters in such a way that the total communication delay will be minimized. Simulation results have shown that the proposed algorithm outperforms the existing algorithms with respect to communication delay and average round-trip time between peers within clusters. Copyright © 2015 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Nowadays, intelligent systems play an important role in a wide range of applications, including financial ones, smart cities, healthcare, and transportation. Most of the intelligent systems are composed of prefabricated components. Inappropriate composition of components may lead to unsafe, power-consuming, and vulnerable intelligent systems. Although artificial intelligence-based systems can provide various advantages for humanity, they have several dark sides that can affect our lives. Some terms, such as security, trust, privacy, safety, and fairness, relate to the dark sides of artificial intelligence, which may be inherent to the intelligent systems. Existing solutions either focus on solving a specific problem or consider the some other challenge without addressing the fundamental issues of artificial intelligence. In other words, there is no general framework to conduct a component selection process while considering the dark sides in the literature. Hence, in this paper, we proposed a new framework for the component selection of intelligent systems while considering the dark sides of artificial intelligence. This framework consists of four phases, namely, component analyzing, extracting criteria and weighting, formulating the problem as multiple knapsacks, and finding components. To the best of our knowledge, this is the first component selection framework to deal with the dark sides of artificial intelligence. We also developed a case study for the component selection issue in autonomous vehicles to demonstrate the application of the proposed framework. Six components along with four criteria (i.e., energy consumption, security, privacy, and complexity) were analyzed and weighted by experts via analytic hierarchy process (AHP) method. The results clearly show that the appropriate composition of components was selected through the proposed framework for the desired functions. [ABSTRACT FROM AUTHOR]