Xian-Bin Cao

Asymmetric cost in snowdrift game on scale-free networks

We study the effects of asymmetric cost on the cooperative behavior in the snowdrift game on scale-free networks. The asymmetric cost reflects the inequality in mutual cooperation and the diversity of cooperators. We focus on the evolution of cooperation and the inequality in wealth distribution influenced by the degree of asymmetry in cost, related with cooperators connections. Interestingly, we find that when cooperators with more neighbors have the advantage, cooperative behavior is highly promoted and the rich exploits the poor to get richer; while if cooperators with less neighbors are favored, cooperation is highly restricted and the rich are forced to offer some payoff to the poor so that the wealth is more homogeneously distributed. The wealth distribution in population is investigated by using the Gini coefficient and the Pareto exponent. Analytical results and discussions are provided to better explain our findings. The asymmetric cost enhances the leader effects in the decision making process by heterogeneous wealth distribution, leading not only to very high cooperator density but also to very stable cooperative behavior.

A real-time schedule method for Aircraft Landing Scheduling problem based on Cellular Automation

The Aircraft Landing Scheduling (ALS) problem has been a complex and challenging problem in air traffic control for a long time. In practice, it can be formulated as a constrained optimization problem that needs to be solved in real-time. Although quite a few optimization techniques, e.g., linear programming-based approaches and evolutionary algorithms, have been shown to be good solver of ALS problems with small number of aircrafts, their relatively high computational cost prohibits their applications in the real world. In this paper, we propose a cellular automata optimization (CAO) approach to the ALS problem. The CAO approach solves the ALS problem in two major steps. First, a good aircraft landing sequence is obtained by simulating the aircraft landing process using a CA model. Then, the exact landing time of each aircraft is determined by a simple yet effective local search procedure. Experimental study on 13 data sets in the OR-Library was conducted to compare the CAO approach and several popular approaches in the literature. It was observed that the CAO method managed to attain high quality solutions on most of the test problem. More importantly, the computational time (in CPU seconds) of CAO method is extremely short. In most cases, satisfactory solutions can be obtained by the CAO approach within 4s, which perfectly fulfills the requirement of the real-world air traffic control system.

Evolutionary Games In Multi-Agent Systems Of Weighted Social Networks

Much empirical evidence has shown realistic networks are weighted. Compared with those on unweighted networks, the dynamics on weighted network often exhibit distinctly different phenomena. In this paper, we investigate the evolutionary game dynamics (prisoners dilemma game and snowdrift game) on a weighted social network consisted of rational agents and focus on the evolution of cooperation in the system. Simulation results show that the cooperation level is strongly affected by the weighted nature of the network. Moreover, the variation of time series has also been investigated. Our work may be helpful in understanding the cooperative behavior in the social systems.

A real-time schedule method for aircraft landing scheduling problem based on cellular automaton

The Aircraft Landing Scheduling (ALS) problem is a typical hard multi-constraint optimization problem. In real applications, it is not most important to find the best solution but to provide a feasible landing schedule in an acceptable time. We propose a novel approach which can effectively solve the ALS while satisfying the real-time need. It consists of two steps: (i) Use CA to simulate the landing process in the terminal airspace and to find a considerably good landing sequence; (ii) a simple Genetic Algorithm associated with a Relaxation Operator is used to obtain a better result based on the CA result. Experiments have shown that our method is much faster and suitable for real-time ALS problem compared with traditional optimization methods. For all the 13 data sets, the proposed approach can find satisfactory solutions in less than 2 seconds.

THE EFFECT OF PINNING CONTROL ON EVOLUTIONARY PRISONER'S DILEMMA GAME

The evolutionary game on graphs provides a natural framework to investigate the cooperation behavior existing in natural and social society. In this paper, degree-based pinning control and random pinning control are introduced into the evolutionary prisoners dilemma game on scale-free networks, and the effects of control mechanism and control cost on the evolution are studied. Numerical simulation shows that forcing some nodes to cooperate (defect) will increase (decrease) the frequency of cooperators. Compared with random pinning control, degree-based pinning control is more efficient, and degree-based pinning control costs less than random pinning control to achieve the same goal. Numerical results also reveal that the evolutionary time series is more stable under pinning control mechanisms, especially under the degree-based pinning control.

Virus-evolutionary genetic algorithm based selective ensemble classifier for pedestrian detection

In pedestrian detection system, it is critical to determine whether a candidate region contains a pedestrian both quickly and reliably. Therefore, an efficient classifier must be designed. In general, a well-organized assembly classifier outperforms than single classifiers. For pedestrian detection, due to the complexity of scene and vast number of candidate regions, an efficient ensemble method is needed.n In this paper, we propose a virus evolutionary genetic algorithm (VEGA) based selective ensemble classifier for pedestrian detection system, in which only part of the trained learners are selected and participate the majority voting for the detection. Component learners are trained with diversity and then VEGA is employed to optimize the selection of component learners. Moreover, a time-spending factor is added to the fitness function so as to balance the detection rate and detection speed. Experiments show that, comparing with typical non-selective Bagging and GA-based selective ensemble method, the VEGA-based selective ensemble gets better performance not only in detecting accuracy but also in detection speed.