基于推理注意力机制的二阶段网络图像修复

现有的图像修复方法在处理大面积缺失或高度纹理化的图像时，通常会产生扭曲的结构或与周围区域不一致的模糊纹理，无法重建合理的图像结构。为此，提出了一种基于推理注意力机制的二阶段网络图像修复方法。首先通过边缘生成网络生成合理的幻觉边缘信息，然后在图像补全网络完成图像的重建工作。为了进一步生成视觉效果更逼真的图像，提高图像修复的精确度，在图像补全网络采用推理注意力机制，有效控制了生成特征的不一致性，从而生成更有效的信息。所提方法在多个数据集上进行了实验验证，结果表明该图像修复方法的结构相似性指数达到了88.9%，峰值信噪比达到了25.56 dB，与现有的图像修复方法相比，该方法具有更高的图像修复精确度，生成的图像更逼真。     

建筑工业化发展的多群体演化博弈与仿真研究

新型建筑工业化具有高质量、低消耗、可循环发展等特征，其推广已上升到国家战略层面。利用演化博弈方法，建立“政府-开发商-银行” 的三方动态演化博弈模型，进行各博弈主体策略的演化稳定性分析，并针对初始状态、奖惩力度、借贷风险和开发成本等对演化结果的影响进行动态仿真。在此基础上， 考虑开发商群体的网络拓扑特征对演化真实性的影响，引入复杂网络理论， 以无标度网络为载体描述开发商个体的连接偏好和决策机制，构建政府监管下的建筑工业化扩散模型，并通过仿真深入研究相关因素对扩散深度的影响作用，最后结合仿真结果给出相应对策建议。     

Leader-following consensus of fractional-order multi-agent systems via adaptive control

This article focuses on the consensus problem of leader-following fractional-order multi-agent systems (MASs) with general linear and Lipschitz nonlinear dynamics. First, the distributed adaptive protocols for linear and nonlinear fractional-order MASs are constructed, respectively. We allow the control coupling gains to be time varying for each agent. Moreover, the adaptive modification schemes for the control gain are designed, which renders smaller control gains and thus requires smaller amplitude on the control input without sacrificing consensus convergence. Second, based on fractional-order Lyapunov stability theorem and Barbalat's lemma, two novel sufficient conditions in terms of linear matrix inequalities are provided to ensure that the leader-following consensus can be obtained in the case for any undirected connected communication graph. Furthermore, we show that the proposed algorithm also works for consensus of agents with intrinsic Lipschitz nonlinear dynamics. As a result, the proposed framework requires no global information and thus can be implemented in a fully distributed manner. Finally, the numerical simulations are given to demonstrate the effectiveness of obtained the theoretical results.     

Controllability and observability of multi-agent systems with heterogeneous and switching topologies

ABSTRACT

This paper focuses on controllability and observability of multi-agent systems with heterogeneous and switching topologies, where the first- and the second-order information interaction topologies are different and switching. First, based on the controllable state set, a controllability criterion is obtained in terms of the controllability matrix corresponding to the switching sequence. Next, by virtue of the subspace sequence, two necessary and sufficient algebraic conditions are established for controllability in terms of the system matrices corresponding to all the possible topologies. Furthermore, controllability is considered from the graphic perspective. It is proved that the system is controllable if the union graph of all the possible topologies is controllable. With respect to observability, two sufficient and necessary conditions are derived by taking advantage of the system matrices and the corresponding invariant subspace, respectively. Finally, some simulation examples are worked out to illustrate the theoretical results.     

Specific emitter identification for satellite communication using probabilistic neural networks

Electromagnetic signal emitted by satellite communication (satcom) transmitters are used to identify specific individual uplink satcom terminals sharing the common transponder in real environment, which is known as specific emitter identification (SEI) that allows for early indications and warning (I&W) of the targets carrying satcom furnishment and furthermore the real time electromagnetic situation awareness in military operations. In this paper, the authors are the first to propose the identification of specific transmitters of satcom by using probabilistic neural networks (PNN) to reach the goal of target recognition. We have been devoted to the examination by exploring the feasibility of utilizing the Hilbert transform to signal preprocessing, applying the discrete wavelet transform to feature extraction, and employing the PNN to perform the classification of stationary signals. There are a total of 1000 sampling time series with binary phase shift keying (BPSK) modulation originated by five types of satcom transmitters in the test. The established PNNs classifier implements the data testing and finally yields satisfactory accuracy at 8 dB(±1 dB) carrier to noise ratio, which indicates the feasibility of our method, and even the keen insight of its application in military.     

When Does Communication Learning Need Hierarchical Multi-Agent Deep Reinforcement Learning

Abstract

Multi-agent systems need to communicate to coordinate a shared task. We show that a recurrent neural network (RNN) can learn a communication protocol for coordination, even if the actions to coordinate are performed steps after the communication phase. We show that a separation of tasks with different temporal scale is necessary for successful learning. We contribute a hierarchical deep reinforcement learning model for multi-agent systems that separates the communication and coordination task from the action picking through a hierarchical policy. We further on show, that a separation of concerns in communication is beneficial but not necessary. As a testbed, we propose the Dungeon Lever Game and we extend the Differentiable Inter-Agent Learning (DIAL) framework. We present and compare results from different model variations on the Dungeon Lever Game.     

Integration of imitation learning using GAIL and reinforcement learning using task-achievement rewards via probabilistic graphical model

The integration of reinforcement learning (RL) and imitation learning (IL) is an important problem that has long been studied in the field of intelligent robotics. RL optimizes policies to maximize the cumulative reward, whereas IL attempts to extract general knowledge about the trajectories demonstrated by experts, i.e, demonstrators. Because each has its own drawbacks, many methods combining them and compensating for each set of drawbacks have been explored thus far. However, many of these methods are heuristic and do not have a solid theoretical basis. This paper presents a new theory for integrating RL and IL by extending the probabilistic graphical model (PGM) framework for RL, control as inference. We develop a new PGM for RL with multiple types of rewards, called probabilistic graphical model for Markov decision processes with multiple optimality emissions (pMDP-MO). Furthermore, we demonstrate that the integrated learning method of RL and IL can be formulated as a probabilistic inference of policies on pMDP-MO by considering the discriminator in generative adversarial imitation learning (GAIL) as an additional optimality emission. We adapt the GAIL and task-achievement reward to our proposed framework, achieving significantly better performance than policies trained with baseline methods.     

The Emotional Dog Gets Mistaken for a Possum.

H. D. Saltzstein and T. Kasachkoff (2004) (see record 2004-21519-004) critique the social intuitionist model (J. Haidt, 2001) (see record 2001-18918-008), but the model they critique is a stripped-down version that should be called the "possum" model. They make 3 charges about the possum model that are not true about the social intuitionist model: that it includes no role for reasoning, that it reduces social influence to compliance, and that it does not take a developmental perspective. After a defense of the honor of the social intuitionist model, this article raises 2 areas of legitimate dispute: the scope and nature of moral reasoning and the usefulness of appealing to innate ideas, rather than to learning and reasoning, as the origin of moral knowledge. This article presents 3 clusters of innate moral intuitions, related to sympathy, hierarchy, and reciprocity. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

条件概率关系数据库模型

现实世界中大量存在着的不确定性信息，关系数据库模型仅视它们为空值，有必要增强其处理这类信息的能力，文章在总结前人工作的基础上推广关系数据库模型，创建有效处理随机型不确定性信息的条件概念关系数据库模型，该模型通过在关系模式中增加一个条件概率测度属性，为每条记录指定适当的条件概率的途径，来表示不确定性信息。文中以对象码为基本工具，创建了条件概率关系结构；以特征函数为基本工具，定义了一套基于该结构的代数运算规则。条件概率的语意比概率的语意广泛，灵活，因而该模型能有效克服概率关系模型的许多不足。