Research of Dynamic Competitive Learning in Neural Networks

Introduce a method of generation of new units within a cluster and aalgorithm of generating new clusters. The model automatically builds up its dynamically growinginternal representation structure during the learning process. Comparing model with other typicalclassification algorithm such as the Kohonen's self-organizing map, the model realizes a multilevelclassification of the input pattern with an optional accuracy and gives a strong support possibilityfor the parallel computational main processor. The idea is suitable for the high-level storage ofcomplex datas structures for object recognition.

STUDIES ON THE DYNAMIC COMPETITIVE ADSORPTION OF ORGANIC VAPORS ON THE ACTIVATED CARBON FIBERS ACTIVATED WITH PHOSPHORIC ACID

The dynamic competitive adsorption behaviors of different binary organic vapor mixtures on ACF-Ps under different operation conditions were investigated by gas chromatography in this paper. The studied mixtures included benzene/toluene, toluene/xylene, benzene/isopropylbenzene, ethyl acetate/toluene and benzene/ethyl acetate. Experimental results show that various ACF-Ps, as with ACF-W, can remove both vapors in binary vapor mixtures with over 99% of removal efficiency before the breakthrough point of the more weakly adsorbed vapor. In dynamic competitive adsorption, the more weakly adsorbed vapor not only penetrates early, but also will be displaced and desorbed consequently by stronger adsorbate and therefore produces a rolling up in the breakthrough curve. The ACF-Ps prepared at different temperatures have somewhat different adsorption selectivity. The feed concentration ratio of vapors, the length/diameter ratio and the thick of bed have effect on competitive adsorption. The competitive adsorption ability of a vapor is mainly related to its boiling point. Usually, the higher the boiling point, the stronger the vapor adsorbed on ACF-P.

Dynamic aggregation evolution of competitive societies of cooperative and noncooperative agents

We propose an evolution model of cooperative agent and noncooperative agent aggregates to investigate the dynamic evolution behaviors of the system and the effects of the competing microscopic reactions on the dynamic evolution. In this model, each cooperative agent and noncooperative agent are endowed with integer values of cooperative spirits and nonco- operative spirits, respectively. The cooperative spirits of a cooperative agent aggregate and the noncooperative spirits of a noncooperative agent aggregate change via four competing microscopic reaction schemes： the win-win reaction between two cooperative agents, the lose-lose reaction between two noncooperative agents, the win-lose reaction between a coop- erative agent and a noncooperative agent （equivalent to the migration of spirits from cooperative agents to noncooperative agents）, and the cooperative agent catalyzed decline of noncooperative spirits. Based on the generalized Smoluchowski＇s rate equation approach, we investigate the dynamic evolution behaviors such as the total cooperative spirits of all coop- erative agents and the total noncooperative spirits of all noncooperative agents. The effects of the three main groups of competition on the dynamic evolution are revealed. These include： （i） the competition between the lose-lose reaction and the win-lose reaction, which gives rise to respectively the decrease and increase in the noncooperative agent spirits; （ii） the competition between the win-win reaction and the win-lose reaction, which gives rise to respectively the increase and decrease in the cooperative agent spirits; （iii） the competition between the win-lose reaction and the catalyzed-decline reaction, which gives rise to respectively the increase and decrease in the noncooperative agent spirits.

A novel chaotic stream cipher and its application to palmprint template protection

Based on a coupled nonlinear dynamic filter （NDF）, a novel chaotic stream cipher is presented in this paper and employed to protect palmprint templates. The chaotic pseudorandom bit generator （PRBG） based on a coupled NDF, which is constructed in an inverse flow, can generate multiple bits at one iteration and satisfy the security requirement of cipher design. Then, the stream cipher is employed to generate cancelable competitive code palmprint biometrics for template protection. The proposed cancelable palmprint authentication system depends on two factors： the palmprint biometric and the password/token. Therefore, the system provides high-confidence and also protects the user＇s privacy. The experimental results of verification on the Hong Kong PolyU Palmprint Database show that the proposed approach has a large template re-issuance ability and the equal error rate can achieve 0.02%. The performance of the palmprint template protection scheme proves the good practicability and security of the proposed stream cipher.

A Study on IPO of CITIC Securities Company

On January 6, 2003, CITIC Securities Co., Ltd. （CITICS hereafter） issued A-shares in the Shanghai Stock Exchange to become the first public listed securities company through IPO in China. CITICS＇s net capital rose to about 5 billion yuan, much higher than any other competitor. This teaching case aims to demonstrate how a company gains first-mover advantage as being the first company to go public in an industry, how its competitors respond by changing their strategies after the IPO, and consequently, how a series of relevant events influence the competition. This case offers important implications for understanding the competitive dynamics of a securities company in China.