最优策略

The theory and method of states evolution are proposed.The agent learns not only the policy but the abstract state space in the evolution process.

提出状态演化的理论和方法，主体在学习最优策略过程中，也学习对环境状态的组织，最终得到互补的抽象状态空间。

On a zero sum two-person continuous game, this paper introduces the concepts of judgement, judgement block, set of optimal strategies under a judgement and so on.

姜殿玉 ，刘广智引进了二人零和连续对策上的判断，判断块以及在判断下的最优策略等概念。

The main contributions are summarized as the following.Firstly, assuming demand is segmented by order lead time, a joint pricing and capacity allocation model is formulated considering scenario-based demand. Due to the stochastic discrete variable in the model, robust optimization is used to solve the model. Numerical examples prove our model and method reflect the stochastic characteristics of demand and provide more pratical pricing and capacity allocation policies than other methods of deterministic and stochastic expectation programming methods especially when the capacity is tight.

本文的创新之处概括为如下几个方面：首先，假设制造企业根据交货提前期对市场细分，建立了考虑需求情境的BTO模式下制造企业定价与能力分配的集成模型，利用稳健优化规划方法对此随机模型进行转化和求解，算例仿真表明随机优化规划方法较确定性规划和随机期望值规划方法更好地反映了随机离散需求的分布和波动特征，更加接近实际运作情况；产能越紧张，各类需求的价格差距越大，三种规划方法给出的最优策略差距越大。

So the conclusion of limited rational faculty analysis can regard as confirming of conclusion and selection tool of completely rational faculty analysis in game theory.

在有限理性的前提下，消费者与经营者的最优策略具有稳定性，使得完全理性的分析结果可以应用于实际中。

Finally, the formula of steganographic capacity under passive attacking or active attacking is described.(3) Under analyzing the maneuver of information hider and attacker in a steganographic system, the sets of strategy for hider and attacker is set up.

从一个新的角度，用对策论的观点研究信息发送者和攻击者行为，给出了构造发送者与攻击者的策略集，进而建立隐秘通信系统对策论模型的方法和求解最优策略的算法。

By space mapping principles, we confirm the relationship between the price curve and the optimal strategy.

我们使用空间映射原理，确定了最优策略与一般价格变化曲线之间的关系。

Markov Decision Process , regarding planning as computing the optimal policy given the model of environment and reward structure, is the most heavily used model.

马尔可夫决策过程是其中应用最多的模型，把规划问题看作是给定环境的模型和奖惩原则的情况下，寻求最优策略的问题。

The ratio of stock right calculated according to venture investment experiential principles is sound and easy to accept. It can give quantitative analysis support to the venture investment practice.(6)According to the theory of the real option, the author derived value formula of the real option fixed price and investment criterion. Based on above, the paper further research for optimal investment strategy of real option in monopolizing and competition under asymmetry information, and built the value model of real option for the investor and administrator by using the theory of the principle trust-agent Meanwhile, the solution of real option optimal investment and the shift value are derived applying the maximal principle .

针对高新技术企业发展的特点：研制周期长、投资费用高、每阶段风险水平特征差别大，运用实物期权理论与计算方法，通过研究建立嵌入期权现金流量图深入分析计算了企业发展中风险投资的最优策略、股权占有比例问题，对不确定性问题由于采用了有关实物期权计算公式，所以计算结果比基于传统的现金流量图的NPV方法更加客观、科学，同时按照风险投资经验法则计算出的股权占有比例合理易于接受，可为企业风险投资实践活动提供数量分析支持。

At first, draws the atomic seeps into executing probabilistic matrix and transferring probabilistic matrix. Change LEG for probability logic seeps through picture of probabilistic logical exploitation graph PLEG, then explains the PLEG problem to the solution of MDP value function and optimal policy.

本文提出了一种基于MDP的风险概率计算模型PLEG-MDP，首先通过引入原子渗透执行概率矩阵和转移概率矩阵，将LEG转换为概率逻辑渗透图PLEG，然后将PLEG解释为MDP，进而求解MDP问题的价值函数和最优策略，其分别对应着目标网络的最大风险概率和风险路径。

Seller should estimate buyer"s behavior and value, choose mechanism and rule, to prevent buyer"s collusion, and achieve maximum of revenue. Buyer has different strategy in four types of auction.

买方在四种经典拍卖方式中的策略和出价行为是不同的，在独自估价模式下，参加英国式拍卖和维克利拍卖的竞买者的最优策略是出价应等于竞买者真正的估价。

Lugalbanda was a god and shepherd king of Uruk where he was worshipped for over a thousand years.

Lugalbanda 是神和被崇拜了一千年多 Uruk古埃及喜克索王朝国王。

I am coming just now,' and went on perfuming himself with Hunut, then he came and sat.

我来只是现在，'歼灭战perfuming自己与胡努特，那麼，他来到和SAT 。