Distributed parallel compilation of MSBNs

Multiply sectioned Bayesian networks (MSBNs) support multiagent probabilistic inference in distributed large problem domains. Inference with MSBNs can be performed using their compiled representations. The compilation involves moralization and triangulation of a set of local graphical structures. Privacy of agents may prevent us from compiling MSBNs at a central location. In earlier work, agents performed compilation sequentially via a depth‐first traversal of the hypertree that organizes local subnets, where communication failure between any two agents would crush the whole work. In this paper, we present an asynchronous compilation method by which multiple agents compile MSBNs in full parallel. Compared with the traversal compilation, the asynchronous one is robust, self‐adaptive, and fault‐tolerant. Experiments show that both methods provide similar quality compilation to simple MSBNs, but the asynchronous one provides much higher quality compilation to complex MSBNs. Empirical study also indicates that the asynchronous one is consistently faster than the traversal one. Copyright © 2009 John Wiley & Sons, Ltd.     

A Bayesian Method for the Induction of Probabilistic Networks from Data

This paper presents a Bayesian method for constructing probabilistic networks from databases. In particular, we focus on constructing Bayesian belief networks. Potential applications include computer-assisted hypothesis testing, automated scientific discovery, and automated construction of probabilistic expert systems. We extend the basic method to handle missing data and hidden (latent) variables. We show how to perform probabilistic inference by averaging over the inferences of multiple belief networks. Results are presented of a preliminary evaluation of an algorithm for constructing a belief network from a database of cases. Finally, we relate the methods in this paper to previous work, and we discuss open problems.     

REPS:一种高效的容错并行概率流Skyline查询方法

概率数据流的并行Skyline查询作为当前大数据分析的一个重要方面,在诸多实际应用中发挥着重要作用。针对并行概率流Skyline查询过程中因发生故障而导致查询结果不准确和查询中断等问题,提出了一种基于复制的容错并行Skyline查询方法REPS。该方法选择参与并行处理的计算节点作为副本节点,并采用层次-循环式数据副本放置策略,选择优先级高的副本恢复数据来保证数据恢复的高效性;同时将故障检测、丢失数据恢复和查询过程恢复贯穿于整个查询更新过程中,以减少容错处理的额外通信和计算开销,并实现快速的容错并行查询。实验结果表明,REPS方法不仅在无故障发生和单个节点失效时具有较高的查询处理效率,而且对于多节点失效情形,仍然能够保持较高的查询处理速率且满足查询需求。     

部分可观测下基于RGMAAC算法的多智能体协同

多智能体深度强化学习(MADRL)将深度强化学习的思想和算法应用到多智能体系统的学习和控制中,是开发具有群智能体的多智能体系统的重要方法.现有的MADRL研究主要基于环境完全可观测或通信资源不受限的假设展开算法设计,然而部分可观测性是多智能体系统实际应用中客观存在的问题,例如智能体的观测范围通常是有限的,可观测的范围外不包括完整的环境信息,从而对多智能体间协同造成困难.鉴于此,针对实际场景中的部分可观测问题,基于集中式训练分布式执行的范式,将深度强化学习算法Actor-Critic扩展到多智能体系统,并增加智能体间的通信信道和门控机制,提出recurrent gated multi-agent Actor-Critic算法(RGMAAC).智能体可以基于历史动作观测记忆序列进行高效的通信交流,最终利用局部观测、历史观测记忆序列以及通过通信信道显式地由其他智能体共享的观察进行行为决策;同时,基于多智能体粒子环境设计多智能体同步且快速到达目标点任务,并分别设计2种奖励值函数和任务场景.实验结果表明,当任务场景中明确出现部分可观测问题时,RGMAAC算法训练后的智能体具有很好的表现,在稳定性...     

Partial abductive inference in Bayesian belief networks using a genetic algorithm

Abductive inference in Bayesian belief networks is the process of generating the K most probable configurations given an observed evidence. When we are only interested in a subset of the network's variables, this problem is called partial abductive inference. Both problems are NP-hard, and so exact computation is not always possible. This paper describes an approximate method based on genetic algorithms to perform partial abductive inference. We have tested the algorithm using the alarm network and from the experimental results we can conclude that the algorithm presented here is a good tool to perform this kind of probabilistic reasoning.     

Learning to communicate in a decentralized environment

Learning to communicate is an emerging challenge in AI research. It is known that agents interacting in decentralized, stochastic environments can benefit from exchanging information. Multi-agent planning generally assumes that agents share a common means of communication; however, in building robust distributed systems it is important to address potential miscoordination resulting from misinterpretation of messages exchanged. This paper lays foundations for studying this problem, examining its properties analytically and empirically in a decision-theoretic context. We establish a formal framework for the problem, and identify a collection of necessary and sufficient properties for decision problems that allow agents to employ probabilistic updating schemes in order to learn how to interpret what others are communicating. Solving the problem optimally is often intractable, but our approach enables agents using different languages to converge upon coordination over time. Our experimental work establishes how these methods perform when applied to problems of varying complexity.     

可控网络中多Agent系统信念可达性和收敛速度分析

在可控网络中,利用多agent系统是网络控制的一种重要方法.在可控网络中,多agent系统中所有agent持有的信念必须在决策前达到与网络实际状态一致,即多agent系统的信念应具有可达性,是实现网络合理控制的基础.传统的基于agent行为的信念更新模型建模过程复杂,不适合在网络控制中直接分析多agent系统信念的可达性和收敛速度.基于传统的信念更新模型,提出了信念距离的概念,并在该概念的基础上提出了新的多agent系统信念距离更新模型,并证明了该模型的合理性.该模型对多agent系统信念距离更新过程加以描述,利用线性系统对多agent系统信念收敛过程进行描述,简化了对多agent系统信念可达性和收敛速度分析的复杂性.在该模型基础上,对网络控制中多agent系统信念可达性和收敛速度进行了分析,给出了判断多agent系统信念可达性的充要条件和收敛速度的上限.另外,针对全耦合网络和无标度网络两种复杂网络的特点,分别对两种网络下多agent系统信念可达性和收敛速度进行了讨论.提出的信念距离更新模型具有良好的适应性,为判断多agent系统的信念可达性提供了有力的工具.     

Cooperating CoScheduling: A Coscheduling Proposal Aimed at Non-Dedicated Heterogeneous NOWs

Implicit coscheduling techniques applied to non-dedicated homogeneous Networks Of Workstations （NOWs） have shown they can perform well when many local users compete with a single parallel job. Implicit coscheduling deals with minimizing the communication waiting time of parallel processes by identifying the processes in need of coscheduling through gathering and analyzing implicit runtime information, basically communication events. Unfortunately, implicit coscheduling techniques do not guarantee the performance of local and parallel jobs, when the number of parallel jobs competing against each other is increased. Thus, a low efficiency use of the idle computational resources is achieved.
In order to solve these problems, a new technique, named Cooperating CoScheduling （CCS）, is presented in this work. Unlike traditional implicit coscheduling techniques, under CCS, each node takes its scheduling decisions from the occurrence of local events, basically communication, memory, Input/Output and CPU, together with foreign events received from cooperating nodes. This allows CCS to provide a social contract based on reserving a percentage of CPU and memory resources to ensure the progress of parallel jobs without disturbing the local users, while coscheduling of communicating tasks is ensured. Besides, the CCS algorithm uses status information from the cooperating nodes to balance the resources across the cluster when necessary. Experimental results in a non-dedicated heterogeneous NOW reveal that CCS allows the idle resources to be exploited efficiently, thus obtaining a satisfactory speedup and provoking an overhead that is imperceptible to the local user.     

多Agent信念逻辑及其在概率意义下的推广

首先建立了一种多Agent信念逻辑MBL(multi-agentbelieflogic),在经典信念逻辑基础上增加了普遍信念算子和公共信念算子,给出MBL的Kripke语义与广义Aumann语义,讨论了两者的等价性,证明了MBL对于上述两种语义的可靠性和完备性.其次,建立了一种多Agent概率信念逻辑MPBL(multi-agentprobabilisticbelieflogic),通过在广义Aumann语义基础上引入概率空间,给出了MPBL的概率Aumann语义,证明了它的可靠性,并给出MPBL的一些推论.     

MAS中一种基于概率论的信任模型

信任是多主体系统（MAS）研究的一个热点问题。为了解决MAS的动态性和不确定性带来的信任问题,提出了一种基于概率论的信任模型。与现有的信任模型相比,该模型考虑了信任信息的完整性和信任的动态性：即在估价agent的信任关系时引入了信任的精确度信心和时间退化因子。模拟实验表明,时间退化因子和信心的引入,能更加有效地评估agent之间的信任关系。     

Thanks for the memory: Cooperative autonomous agent search in uncertain environments

The effects of cooperation between autonomous electronic or physical agents are widely studied in computational science literature. We concentrate on a homogenous population of agents in a multi-agent system (MAS) to explore the effects of useful memory on goal achievement. We use simulations to consider two-dimensional planar surfaces upon which N targets are randomly scattered. N agents exist each with a maximal interest in one specific target. Agents may observe the positions of “uninteresting” targets in the environment and communicate this information to other agents encountered within the environment. The benefits of cooperation can be approximated by pure probabilistic analysis for theoretical search success, but the introduction of real-world cost factors (e.g. fuel, energy, transmission time) associated with movement within the environment renders these predictions unusable. In pure probabilistic terms, higher numbers of cooperative agents can greatly increase search effectiveness. In systems where positive costs are associated with search, internal agent memory factors can allow agent density to approximate pure probabilistic effectiveness. Practical applications for this research include real-time electronic document search, problems in robotic multi-agent systems (e.g. “foraging” or “consumption” problems), and network coverage for wireless communication devices.     

Variations over the message computation algorithm of lazy propagation.

Improving the performance of belief updating becomes increasingly important as real-world Bayesian networks continue to grow larger and more complex. In this paper, an investigation is done on how variations over the message-computation algorithm of lazy propagation may impact its performance. Lazy propagation is a junction-tree-based inference algorithm for belief updating in Bayesian networks. Lazy propagation combines variable elimination (VE) with a Shenoy-Shafer message-passing scheme in an attempt to exploit the independence properties induced by evidence in a junction-tree-based algorithm. The authors investigate, the use of arc reversal (AR) and symbolic probabilistic inference (SPI) as alternative algorithms for computing clique-to-clique messages in lazy propagation. The paper presents the results of an empirical evaluation of the performance of lazy propagation using AR, SPI, and VE as the message-computation algorithm. The results of the empirical evaluation show that no single algorithm outperforms or is outperformed by the other two alternatives. In many cases, there is no significant difference in the performance of the three algorithms.     

Convergence analysis of max-consensus algorithm in probabilistic communication networks with Bernoulli dropouts

In the presence of probabilistic communication networks between agents, the convergence analysis of max-consensus algorithm (MCA) is addressed in this paper. It is considered that at each iteration of MCA, all agents share their measurements with adjacent agents via local communication networks which is applicable in many multi-agent systems (MASs). It is assumed that the communication networks have Bernoulli dropouts, i.e. the information exchanged between agents may be lost with Bernoulli distribution. In the proposed method, the information topology of MAS is modelled as a dynamic graph with the Bernoulli adjacency matrix. It is proved that in the presence of Bernoulli dropouts and under non-restrictive assumptions concerning the MAS features and communication topology, the MCA converges with a probability one in the finite time. Furthermore, the upper bounds are provided by means of deterministic and probabilistic expressions for the expectation and dispersion of convergence time, respectively. It is shown that the proposed upper bounds are asymptotic, i.e. there are specific conditions of MAS in which the convergence time of MCA tends to the proposed upper bounds. The convergence accuracy of MCA is discussed in terms of probabilistic equations. The validity of the proposed theorems is illustrated by means of simulation results.     

一种信度网推理的并行计算模型

基于概率知识表达的信度网已成为人工智能中非确定知识表达和推理的研究热点。推理算法是信度网学习和应用的基础。该文提出了一种基于经典Polytree算法的推理计算模型。该模型表达清楚,计算过程容易控制,并能够简单地映射到并行机结构上。该文首先介绍了模型在单联通网络下的计算步骤,然后将模型引入到多联通网络上。     

Using communication to reduce locality in distributed multiagent learning

Abstract. This paper attempts to bridge the fields of machine learning, robotics, and distributed AI. It discusses the use of communication in reducing the undesirable effects of locality in fully distributed multi-agent systems with multiple agents robots learning in parallel while interacting with each other. Two key problems, hidden state and credit assignment, are addressed by applying local undirected broadcast communication in a dual role: as sensing and as reinforcement. The methodology is demonstrated on two multi-robot learning experiments. The first describes learning a tightly-coupled coordination task with two robots, the second a loosely-coupled task with four robots learning social rules. Communication is used to (1) share sensory data to overcome hidden state and (2) share reinforcement to overcome the credit assignment problem between the agents and bridge the gap between local individual and global group pay-off.     

Data Parallel Implementation of Belief Propagation in Factor Graphs on Multi-core Platforms

We investigate data parallel techniques for belief propagation in acyclic factor graphs on multi-core systems. Belief propagation is a key inference algorithm in factor graph, a probabilistic graphical model that has found applications in many domains. In this paper, we explore data parallelism for basic operations over the potential tables in belief propagation. Data parallel techniques for these table operations are developed for shared memory platforms. We then propose a complete belief propagation algorithm using these table operations to perform exact inference in factor graphs. The proposed algorithms are implemented on state-of-the-art multi-socket multi-core systems with additional NUMA-aware optimizations. Our proposed algorithms exhibit good scalability using a representative set of factor graphs. On a four-socket Intel Westmere-EX system with 40 cores, we achieve 39.5 $\times $ speedup for the table operations and 39 $\times $ speedup for the complete algorithm using factor graphs with large potential tables.     

Propositional Attitudes and Causation

Predicting and explaining the behavior of others in terms of mental states is indispensable for everyday life. It will be equally important for artificial agents. We present an inference system for representing and reasoning about mental states, and use it to provide a formal analysis of the false-belief task. The system allows for the representation of information about events, causation, and perceptual, doxastic, and epistemic states (vision,belief, and knowledge), incorporating ideas from the event calculus and multi-agent epistemic logic. Unlike previous AI formalisms, our focus here is on mechanized proofs and proof programmability, not on metamathematical results. Reasoning is performed via relatively cognitively plausible inference rules, and a degree of automation is achieved by generalpurpose inference methods and by a syntactic embedding of the system in first-order logic.     

多agent系统的一种交互策略模型

在多agent系统(MAS)中,通信交互是agent实现协作的主要途径.文章从语义层的角度对agent间通信交互过程进行分析,将agent的思维状态BDI(belief,desire,intention)模型引入通信交互过程,提出一种交互策略模型,支持在基本交互行为之上的多种类型的协商交互,以解决agent间的信知、行动等方面的冲突.与以往的研究中的辩论协商等方法相比,该策略模型可以实现基于场景的灵活交互,更具实用性.     

基于后悔值的多Agent冲突博弈强化学习模型

对于冲突博弈,研究了一种理性保守的行为选择方法,即最小化最坏情况下Agent的后悔值.在该方法下,Agent当前的行为策略在未来可能造成的损失最小,并且在没有任何其他Agent信息的条件下,能够得到Nash均衡混合策略.基于后悔值提出了多Agent复杂环境下冲突博弈的强化学习模型以及算法实现.该模型中通过引入交叉熵距离建立信念更新过程,进一步优化了冲突博弈时的行为选择策略.基于Markov重复博弈模型验证了算法的收敛性,分析了信念与最优策略的关系.此外,与MMDP(multi-agent markov decision process)下Q学习扩展算法相比,该算法在很大程度上减少了冲突发生的次数,增强了Agent行为的协调性,并且提高了系统的性能,有利于维持系统的稳定.     

Verification of heterogeneous multi-agent system using MCMAS

The focus of the paper is how to model autonomous behaviours of heterogeneous multi-agent systems such that it can be verified that they will always operate within predefined mission requirements and constraints. This is done by using formal methods with an abstraction of the behaviours modelling and model checking for their verification. Three case studies are presented to verify the decision-making behaviours of heterogeneous multi-agent system using a convoy mission scenario. The multi-agent system in a case study has been extended by increasing the number of agents and function complexity gradually. For automatic verification, model checker for multi-agent systems (MCMAS) is adopted due to its novel capability to accommodate the multi-agent system and successfully verifies the targeting behaviours of the team-level autonomous systems. The verification results help retrospectively the design of decision-making algorithms improved by considering additional agents and behaviours during three steps of scenario modification. Consequently, the last scenario deals with the system composed of a ground control system, two unmanned aerial vehicles, and four unmanned ground vehicles with fault-tolerant and communication relay capabilities.