An intercausal cancellation model for Bayesian-network engineering

When constructing Bayesian networks with domain experts, network engineers often use the noisy-OR model, and causal interaction models more generally, to alleviate the burden of probability elicitation: the use of such a model serves to reduce the number of probabilities to be elicited on the one hand, and on the other hand forestalls experts having to give assessments for probabilities with compound conditions which they feel are hard to envision. Recently, we have shown that ill-considered use of the noisy-OR model specifically can substantially decrease a network's performance, especially in domains in which causal mechanisms include cancellation effects. Motivated by this observation, we designed a new causal interaction model, with the same engineering advantages as the noisy-OR model, to describe such effects. We detail properties of our intercausal cancellation model, and compare it against existing causal interaction models. We further illustrate the application of our model in the real-world domain of pharmacology.     

Marginal filtering in large state spaces

Recognising everyday activities including information about the context requires to handle large state spaces. The usage of wearable sensors like six degree of freedom accelerometers increases complexity even more. Common approaches are unable to maintain an accurate belief state within such complex domains. We show how marginal filtering can overcome limitations of standard particle filtering and efficiently infer the context of actions. Symbolic models of human behaviour are used to recognise activities in two different settings with different state space sizes. Based on these scenarios we compare the marginal filter to the standard particle filter. An evaluation shows that the marginal filter performs comparably in small state spaces but outperforms the particle filter in large state spaces.     

时变偏微分方程的贝叶斯稀疏识别方法

在数据驱动的建模中,通过测量或模拟得到时空数据,我们发现基于拉普拉斯先验的贝叶斯稀疏识别方法能有效地恢复时变偏微分方程的稀疏系数.本文将贝叶斯稀疏识别方法运用于各种时变偏微分方程模型(KdV方程、Burgers方程、Kuramoto-Sivashinsky方程、反应-扩散方程、非线性薛定谔方程和纳维-斯托克斯方程)的方...     

贝叶斯优化卷积神经网络公共场所异常声识别*

针对公共场所异常声的感知和识别问题,提出一种基于贝叶斯优化卷积神经网络的识别方法。提取声信号的Gammatone倒谱系数、倍频程功率谱、短时能量和谱质心,组合成声信号的特征图。构建卷积神经网络作为分类器,利用递增的卷积核设置和池化操作处理不同尺度的特征。基于贝叶斯优化算法优化卷积神经网络的模型参数,对包括火苗噼啪声、婴儿啼哭声、烟花燃放声、玻璃破碎声和警报声的5种公共场所异常声进行识别。该方法的识别结果与基于不同的特征提取和分类器方案得到的识别结果进行比较,结果表明该方法的识别效果优于其他特征提取和分类器方案的识别效果。最后分析了该方法在不同信噪比噪声干扰下的识别结果,验证了该方法的有效性。     

基于贝叶斯网络的船舶港口国监督检查滞留风险研究

针对港口国监督(Port State Control, PSC)检查的复杂性和不确定性, 基于贝叶斯网络理论构建船舶PSC检查滞留风险分析模型. 以东京备忘录(Tokyo MOU)中2014~2017年船舶PSC检查样本数据为基础, 运用R语言bnlearn包进行贝叶斯网络的结构及参数学习. 同时分别执行贝叶斯网络的正向、逆向推理, 定量表示各风险因素与滞留结果之间的相互作用关系, 找出导致船舶滞留的高风险因素, 实现不确定环境下船舶PSC检查滞留风险的全面动态分析. 实证表明, 模型具有较高的精确度, 可为检查人员的滞留决策及航运公司的安全风险管理提供有效依据.     

基于显示连通贝叶斯网络的机场塔台地震易损性分析

将基于性能的多维易损性分析方法,结合显示连通贝叶斯网络,应用于机场塔台的多维易损性分析。考虑地震激励的不确定性,通过非线性时程分析获得结构响应数据;将塔台结构分为三个层次,每个层次按包含的层数分为相应的子层次。根据功能特性确定子层次的评价指标和极限状态,建立服从多元对数正态分布的概率地震需求模型;考虑各种极限状态之间的相关性,建立极限状态方程,确定失效域,通过蒙特卡洛法求得构件的超越概率;建立塔台结构的显示连通贝叶斯网络模型,利用层次分析法获得中间节点的条件概率表,利用MATLAB进行贝叶斯网络的推理计算,实现从单一层次的易损性到整体易损性的推理。     

AN ACCELERATION STRATEGY FOR RANDOMIZE-THEN-OPTIMIZE SAMPLING VIA DEEP NEURAL NETWORKS

Randomize-then-optimize (RTO) is widely used for sampling from posterior distribu-tions in Bayesian inverse problems.However,RTO can be computationally intensive for complexity problems due to repetitive evaluations of the expensive forward model and its gradient.In this work,we present a novel goal-oriented deep neural networks (DNN) sur-rogate approach to substantially reduce the computation burden of RTO.In particular,we propose to drawn the training points for the DNN-surrogate from a local approximated posterior distribution-yielding a flexible and efficient sampling algorithm that converges to the direct RTO approach.We present a Bayesian inverse problem governed by elliptic PDEs to demonstrate the computational accuracy and efficiency of our DNN-RTO ap-proach,which shows that DNN-RTO can significantly outperform the traditional RTO.     

Chess AI: Competing Paradigms for Machine Intelligence

Endgame studies have long served as a tool for testing human creativity and intelligence. We find that they can serve as a tool for testing machine ability as well. Two of the leading chess engines, Stockfish and Leela Chess Zero (LCZero), employ significantly different methods during play. We use Plaskett’s Puzzle, a famous endgame study from the late 1970s, to compare the two engines. Our experiments show that Stockfish outperforms LCZero on the puzzle. We examine the algorithmic differences between the engines and use our observations as a basis for carefully interpreting the test results. Drawing inspiration from how humans solve chess problems, we ask whether machines can possess a form of imagination. On the theoretical side, we describe how Bellman’s equation may be applied to optimize the probability of winning. To conclude, we discuss the implications of our work on artificial intelligence (AI) and artificial general intelligence (AGI), suggesting possible avenues for future research.     

基于贝叶斯参数优化的无模型自适应硅单晶直径控制

直拉硅单晶的生长过程涉及多场多相耦合与复杂的物理化学变化,其中工艺参数的波动是导致晶体直径不均匀的重要原因,如何实现工艺参数的控制以获得理想的、均匀的晶体直径具有重要的研究意义。本文分析现有控制方法存在不稳定以及控制效果不佳的问题后,提出基于贝叶斯参数优化的无模型自适应控制模型来控制硅单晶生长过程中的晶体直径。首先以坩埚上升速度与加热器的功率作为控制输入参数,晶体直径作为输出,搭建无模型自适应控制模型,并分析算法的稳定性。其次将控制模型进行仿真实验,发现硅单晶直径控制模型中不同的超参数设定会影响控制过程的迭代次数以及控制效果。最后,利用贝叶斯优化超参数的取值范围,并进行最终的仿真实验,结果表明,经贝叶斯参数优化后的控制模型计算快、迭代次数少,输出的晶体直径稳定,同时将生长工艺参数控制在实际生产要求范围内。因此,基于贝叶斯参数优化的无模型自适应控制实现了硅单晶直径均匀稳定的有效控制,具有结合工程背景的实际应用前景。