导弹发射决策专家系统推理机研究与实现

介绍了一种基于灵活、通用的知识表达方法的推理机。它将面向对象方法和问题求解的黑板模型有机结合 ,自带各种控制信息和适宜的不确定性推理算法 ,将知识库分层和模块化 ,极大地提高了推理效率。推理机根据监控信息 ,推理计算出导弹允许发射的合格置信度 ,已应用于导弹发射决策专家系统设计中。     

Inference to the hypothesis of extended cognition

This paper examines the justification for the hypothesis of extended cognition (HEC). HEC claims that human cognitive processes can, and often do, extend outside our head to include objects in the environment. HEC has been justified by inference to the best explanation (IBE). Both advocates and critics of HEC claim that we can infer the truth value of HEC based on whether HEC makes a positive or negative explanatory contribution to cognitive science. I argue that IBE cannot play this epistemic role. A serious rival to HEC exists with a differing truth value, and this invalidates IBEs for both the truth and the falsity of HEC. Explanatory value to cognitive science is not a guide to the truth value of HEC.     

论CAPP的创成模式

本文论述了系统运用成组技术原理，提出了检索，创成综合模式。其特点是按相似零件组编组工序、工步词典，以工序工步词典为索引建立相应的工艺决策规则。在系统软件实现方面，运用编译原理、构造词法、语法分析器，对规则或知识进行预处理，形成特定的语法结构，使知识库与推理机相互独立。     

从“附注”管窥贝叶斯推理

资料显示,来自Fisher的批评以及诸如皮尔森、杰弗里斯等的论证已经曲解了作为非充分性理由要求的贝叶斯论证。重新审视托马斯·贝叶斯的著名附注(注释)中他为未知概率的先验均匀分布给出的辩护理由,发现被指控的原理性缺陷,在原贝叶斯文献中是不存在的。研究认为,真正的"贝叶斯推理"与早期的批评者和评论者对之的转述相比较差异很大,在逻辑上更接近于现代意义上对该概念的理解。     

Beyond heuristics and biases: A contingency model of judgemental forecasting

The conflicting viewpoints about the quality of judgemental forecasts are examined and a model is proposed that attempts to resolve the conflict. The model sees forecasts as contingent upon the repertory of forecasting strategies that the forecaster brings to the forecasting task, the strategy that he or she selects as a function of the characteristics of the task, and the rigour with which he or she applies the strategy as a function of the motivating characteristics of the environment in which the task is encountered. The implications of differences in subjects' and experimenters' assumptions about which strategies are appropriate in experimental studies are examined, as are the implications of the differences between the motivating aspects of experimental and applied settings on both performance and on the generatizability of the results of experiments to applied judgemental forecasting.     

基于灰色PSO-ANFIS模型的船舶横摇运动预报分析

为了准确高效的预测船舶在波浪中的航行状态以保证人员、货物和船舶的安全,提出了一种基于灰色模型粒子群优化算法的自适应神经模糊推理系统(Grey Particle Swarm Optimization-Adaptive Neural-fuzzy Inference System, GPSO -ANFIS)。GPSO-ANFIS预测模型使用模糊C均值聚类算法对输入样本进行聚类分析,得到模糊规则数量并建立神经模糊推理系统,再使用粒子群优化算法对建立的预测系统进行优化训练,从而得到最优的预测系统模型。其中灰色模型用于横摇数据的预处理,以便削弱横摇状态中的非线性影响因素。最后通过实船“育鲲”轮的横摇数据进行仿真实验,实验结果验证了GPSO-ANFIS模型的实用性和可行性,具有较高的预测精度。并为船舶航行智能化提供了一种有价值的理论依据。     

Exemplification and the use-values of cases and case studies

This paper provides an account of the ‘use-value’ of case-based research by showing how social scientists exploit cases, and case studies, in a variety of practices of inference and extension. The critical basis for making such extensions relies on the power of a case, or the account given of a case (the case-study account), to exemplify certain features of the social world in ways which prove valuable for further analysis: either of the same case, or in many domains beyond the original case study. Framing use-values in terms of exemplification compares favourably with understanding reasoning beyond the case either as a form of analogical reasoning or in taking cases as experimentable objects.     

The ecological rationality of explanatory reasoning

There is growing evidence that explanatory considerations influence how people change their degrees of belief in light of new information. Recent studies indicate that this influence is systematic and may result from people’s following a probabilistic update rule. While formally very similar to Bayes’ rule, the rule or rules people appear to follow are different from, and inconsistent with, that better-known update rule. This raises the question of the normative status of those updating procedures. Is the role explanation plays in people’s updating their degrees of belief a bias? Or are people right to update on the basis of explanatory considerations, in that this offers benefits that could not be had otherwise? Various philosophers have argued that any reasoning at deviance with Bayesian principles is to be rejected, and so explanatory reasoning, insofar as it deviates from Bayes’ rule, can only be fallacious. We challenge this claim by showing how the kind of explanation-based update rules to which people seem to adhere make it easier to strike the best balance between being fast learners and being accurate learners. Borrowing from the literature on ecological rationality, we argue that what counts as the best balance is intrinsically context-sensitive, and that a main advantage of explanatory update rules is that, unlike Bayes’ rule, they have an adjustable parameter which can be fine-tuned per context. The main methodology to be used is agent-based optimization, which also allows us to take an evolutionary perspective on explanatory reasoning.     

基于多层次动态门控推理网络的文本蕴含识别

现有的文本蕴含模型通常计算一次词级别注意力得到两段文本在不同层面的交互特征,但对于文本不同层面的理解,不同重要词的注意力应该是不同的,并且一次词级注意力推理仅能捕捉到文本对局部特征.针对这个问题,提出一种多层次动态门控推理网络,该网络结合了词级别信息的细粒度推理和句子级别门控机制来动态捕捉文本对的语义信息,并采用不同注意力计算方式提取文本对不同层面的语义特征,共同推理文本对的蕴含关系.本文在两个文本蕴含数据集上均做了实验,相较于基准模型和现有主流模型,准确率提升了0.4%～1.7%,通过消融分析,进一步验证了本文模型各部分结构的有效性.