Corrective feedback control of competing neural network with entire connections

Continuous persist activity of the competitive network is related to many functions, such as working memory, oculomotor integrator and decision making. Many competition models with mutual inhibition structures achieve activity maintenance via positive feedback, which requires meticulous fine tuning of the network parameters strictly. Negative derivative feedback, according to recent research, might represent a novel mechanism for sustaining neural activity that is more resistant to multiple neural perturbations than positive feedback. Many classic models with only mutual inhibition structure are not capable of providing negative derivative feedback because double-inhibition acts as a positive feedback loop, and lack of negative feedback loop that is indispensable for negative derivative feedback. Here in the proposal, we aim to derive a new competition network with negative derivative feedback. The network is made up of two symmetric pairs of EI populations that the four population are completely connected. We conclude that the negative derivative occurs in two circumstances, in which one the activity of the two sides is synchronous but push-pull-like in the other, as well as the switch of two conditions in mathematical analysis and numerical simulation.

Network structure of resource use and niche overlap within the endophytic microbiome

Endophytes often have dramatic effects on their host plants. Characterizing the relationships among members of these communities has focused on identifying the effects of single microbes on their host, but has generally overlooked interactions among the myriad microbes in natural communities as well as potential higher-order interactions. Network analyses offer a powerful means for characterizing patterns of interaction among microbial members of the phytobiome that may be crucial to mediating its assembly and function. We sampled twelve endophytic communities, comparing patterns of niche overlap between coexisting bacteria and fungi to evaluate the effect of nutrient supplementation on local and global competitive network structure. We found that, despite differences in the degree distribution, there were few significant differences in the global network structure of niche-overlap networks following persistent nutrient amendment. Likewise, we found idiosyncratic and weak evidence for higher-order interactions regardless of nutrient treatment. This work provides a first-time characterization of niche-overlap network structure in endophytic communities and serves as a framework for higher-resolution analyses of microbial interaction networks as a consequence and a cause of ecological variation in microbiome function.

Optimization and improvement of a robotics gaze control system using LSTM networks

Gaze control represents an important issue in the interaction between a robot and humans. Specifically, deciding who to pay attention to in a multi-party conversation is one way to improve the naturalness of a robot in human-robot interaction. This control can be carried out by means of two different models that receive the stimuli produced by the participants in an interaction, either an on-center off-surround competitive network or a recurrent neural network. A system based on a competitive neural network is able to decide who to look at with a smooth transition in the focus of attention when significant changes in stimuli occur. An important aspect in this process is the configuration of the different parameters of such neural network. The weights of the different stimuli have to be computed to achieve human-like behavior. This article explains how these weights can be obtained by solving an optimization problem. In addition, a new model using a recurrent neural network with LSTM layers is presented. This model uses the same set of stimuli but does not require its weighting. This new model is easier to train, avoiding manual configurations, and offers promising results in robot gaze control. The experiments carried out and some results are also presented.

Distributed multiple-bipartite consensus in networked Lagrangian systems with cooperative-competitive interactions

In combination with the collective behavior evolutions of bipartite consensus and cluster/group consensus, this paper proposes the notion of multiple-bipartite consensus in networked Lagrangian systems (NLSs). The distributed leaderless and leader-following multiple-bipartite consensus control laws for NLSs are presented in the cooperative-competitive network, where the negative interactions between agents can exist in the same subnetwork. By introducing an acyclic partition and adding the integral item in the control protocols, the final explicit convergence states in the leaderless case are eventually obtained. Moreover, the leader-following scenario can be realized in fifinite time with integrated controllers. All of the effectiveness has been illustrated through numerical simulations.

Optimization and Decision of Supply Chain Considering Negative Spillover Effect and Service Competition

Excellent service plays a vital role in the sustainability of enterprise and supply chains development in today’s increasingly fierce market competition. However, due to the inevitable spillover effect in the competitive network, enterprises’ initiative to improve the service level is reduced. From the perspective of negative spillover effect, optimization and decision-making in the competitive network of retailer-dominated supply chain are examined in this study. Considering four competitive situations in practical operation management, the corresponding double-layer compound nested Stackelberg game models are constructed, and the optimal equilibrium solutions are derived. Employing comprehensive comparison and analysis of the results, it is found that when the negative spillover effect of service increases, the optimal profit and service level of the leading supply chain or its retailers decrease, and the optimal retail price and overall optimal profit also gradually decline. For the leading supply chain, the centralized decision-making can achieve higher profits, and also more willing to improve the level of service. However, for the following supply chain, when the negative spillover effect of service is weak, the optimal service level under decentralized decision is higher, while when the spillover effect of service is strong, the optimal service level under integrated decision is higher. In addition, the supply chain-to-chain competition can bring negative incentives to the retailer that provides services, while for the rival that does not provide services, it can generate a certain free-riding effect that benefits them, and the effect is enhanced with the increase of competition.

Bipartite Consensus Control for a Swarm of Robots

This paper studies the bipartite consensus problem of a swarm of robots whose dynamics are formulated by Lagrangian equations. Two distributed bipartite consensus control protocols are proposed for a swarm of robots without a leader or with a virtual leader. For the nonleader case, the networked Lagrangian system can reach static bipartite consensus under the control protocol developed, and the final convergent states can be explicitly determined by the specific structure of the Laplacian matrix associated with the cooperative–competitive network topology. For the virtual leader case, all the followers can track the leader's state in a bipartite formation to realize bipartite tracking consensus. Finally, the simulation results are given to verify the theoretical results.

Cooperative Multi-Agent Interaction and Evaluation Framework Considering Competitive Networks with Dynamic Topology Changes

In recent years, the problem of reinforcement learning has become increasingly complex, and the computational demands with respect to such processes have increased. Accordingly, various methods for effective learning have been proposed. With the help of humans, the learning object can learn more accurately and quickly to maximize the reward. However, the rewards calculated by the system and via human intervention (that make up the learning environment) differ and must be used accordingly. In this paper, we propose a framework for learning the problems of competitive network topologies, wherein the environment dynamically changes agent, by computing the rewards via the system and via human evaluation. The proposed method is adaptively updated with the rewards calculated via human evaluation, making it more stable and reducing the penalty incurred while learning. It also ensures learning accuracy, including rewards generated from complex network topology consisting of multiple agents. The proposed framework contributes to fast training process using multi-agent cooperation. By implementing these methods as software programs, this study performs numerical analysis to demonstrate the effectiveness of the adaptive evaluation framework applied to the competitive network problem depicting the dynamic environmental topology changes proposed herein. As per the numerical experiments, the greater is the human intervention, the better is the learning performance with the proposed framework.

CAPTCHA recognition based on a generative-adversarial network

CAPTCHA recognition is certainly not a new research topic. Over the past decade, researchers have demonstrated various ways to automatically recognize text-based CAPTCHAs. However, in such methods, the recognition setup requires a large participation of experts and carries a laborious process of collecting and marking data. This article presents a general, low-cost, but effective approach to automatically solving text-based CAPTCHAs based on deep learning. This approach is based on the architecture of a generative-competitive network, which will significantly reduce the number of real required CAPTCHAs.

An Extension of Fuzzy Competition Graph and Its Uses in Manufacturing Industries

Competition graph is a graph which constitutes from a directed graph (digraph) with an edge between two vertices if they have some common preys in the digraph. Moreover, Fuzzy competition graph (briefly, FCG) is the higher extension of the crisp competition graph by assigning fuzzy value to each vertex and edge. Also, Interval-valued FCG (briefly, IVFCG) is another higher extension of fuzzy competition graph by taking each fuzzy value as a sub-interval of the interval [ 0 , 1 ] . This graph arises in many real world systems; one of them is discussed as follows: Each and every species in nature basically needs ecological balance to survive. The existing species depends on one another for food. If there happens any extinction of any species, there must be a crisis of food among those species which depend on that extinct species. The height of food crisis among those species varies according to their ecological status, environment and encompassing atmosphere. So, the prey to prey relationship among the species cannot be assessed exactly. Therefore, the assessment of competition of species is vague or shadowy. Motivated from this idea, in this paper IVFCG is introduced and several properties of IVFCG and its two variants interval-valued fuzzy k-competition graphs (briefly, IVFKCG) and interval-valued fuzzy m-step competition graphs (briefly, IVFMCG) are presented. The work is helpful to assess the strength of competition among competitors in the field of competitive network system. Furthermore, homomorphic and isomorphic properties of IVFCG are also discussed. Finally, an appropriate application of IVFCG in the competition among the production companies in market is presented to highlight the relevance of IVFCG.