基于随机事件处理的情感建模研究

针对各种传感器采集到的随机数字量，进行归一化和修正，使之符合人类对外界刺激的感觉。同时利用人工心理理论建立了一个基于随机过程的情感空间，用修正过的外界刺激作为输入，产生模拟的人类情感。这种产生的模拟的情感同时也对模型参数进行修正。     

Enacting archetypes in movies: grounding the unconscious mind in emotion-driven media

Abstract

The concept of emotion-driven media integrates affective computing into developing new types of digital media by including the human spectator as an essential part of the whole system. Enactive media, as a landmark in this direction, provides a theoretical basis that is rooted in the enactivist approach in cognitive science. Based on this thesis, it is required to establish a practical procedure for developing the media content through integrating psychological approaches. We demonstrate a systematic approach for developing the content for emotion-driven media. The theory of archetypes, one of the psychoanalytical approaches, was applied for editing movie clips as the media content, and coupled the content with emotional responses by using experimental psychological methodologies. The results showed that it is promising to utilise archetypal movie clips as the content of emotion-driven media and archetypal symbolism can be a useful resource for developing the content of media systems.     

基于个性和OCC的机器人情感建模研究

机器人不仅要具有简单的机械作业和逻辑推理能力，还应当具有类似人类的情感能力．本文将个性与情绪、情感、理解、表达相结合，采用OCC模型作为评价标准，建立了符合人类情感规律的、可用于情感机器人的情感模型。通过一个应用上述模型的虚拟人情感交互系统．验证了此模型可以很好的对人类的情感进行仿真．可以应用于情感机器人和人性化计算机、游戏等许多领域。     

一类面向多目标决策的分层情感认知模型

决策者的情感认知对于复杂多目标决策具有关键的作用, 为了能使计算机掌握决策者的决策情感认知规律, 提升计算机辅助多目标决策的能力, 提出一种融合了性格特征、心境状态和情感状态的多层情感认知模型. 描述决策者在处理复杂决策问题时的情感状态量化关系, 运用分层情感认知模型构建集感性和理性于一身的虚拟人, 代替决策者完成对多目标优化问题的交互决策求解. 实例分析验证了所提出模型的有效性.

     

情感建模与情感识别

情感计算是关于、产生于和影响于情感方面的计算,其目的是赋予计算机识别、理解、表达和适应人情感的能力。情感计算通过各种传感器获取由人的情感所引起的表情及生理变化信号,利用“情感模型”对这些信号进行识别,从而理解人的情感并做出适当的响应。该文主要讨论了Picard教授在情感计算中情感识别部分的研究成果,着重分析了面部表情、语音、生理信号的情感模型与情感识别,这是情感计算研究的一个关键问题之一,也是建立和谐人机环境的基础之一。     

采用Adaboost算法进行面部表情识别

Adaboost是一种有效的分类器组合方法，它能够提高弱分类器的分类性能。利用Adaboost方法对面部表情进行识别，探讨了Adaboost与主成分分析法的几种结合方案。仿真结果表明这几种方案可行，且识别效果较好。     

人机交互中的个性化情感模型

人与机器人的交互过程中,情感因素的引入能够使人机交流更加自然和谐.因此,完整的人工情感模型的建立是首要解决的问题.基于情感能量理论基础,首先,提出了心境自发转移和刺激转移模型.其次,结合情绪自发转移的马尔可夫链模型和刺激转移的HMM模型,将心境和情绪的自发和刺激转移过程统一在一个框架下.最后,将完整的人工情感模型软件化并应用于儿童玩伴机器人上,在接受非结构化环境与用户的信息输入后,个性化的情感软件模块产生输出,实现针对儿童用户的玩伴机器人个性化交互,通过应用验证了该模型的有效性.     

Embodied imagination: a hybrid method of designing for intimacy

Abstract

Situated in the domain of research into mobile, wireless, networked and wearable computing, this exploratory paperintroduces the embodied imagination method and explains how it can contribute to the design process by creating an elastic space of performance that incorporates daily life and personal imagination into the design process. It is based on a study called Placebo Sleeves which was an experiential design phase of a larger project in wearable computing called whisper[s]. The innovation offered by this research is twofold: an integration of previously distinct methodologies, and an interdisciplinary theoretical framework relevant to the design of devices for affective, networked communication. The methodologies are shaped both by user experience models and by performance practices. We also articulate a domain of public dreaming, located at the conjunction of the private, public and secret within human existence, and suggest that shared use of mobile technologies has the potential to be situated there.     

The Etienne Grandjean Memorial Lecture Situated cognition and action: implications for ergonomic work analysis and anthropotechnology

Abstract

After a short history of the study of work activities, the methodology of Ergonomic Work Analysis (EWA) is described and the diversity of practices is underlined. The methodology of analysis of activities involves exhaustive checks of the behaviour of operators in critical situations and confronts the operator with his own behaviour in order to obtain pertinent explanations and evoke the cognitive unconscious. Ethnological work may constitute a contribution as regards the choice of the operator(s) whose behaviour is the most significant for the problem posed. In the same way, ethnologists using recording tools that are similar to those of ergonomists offer interesting frameworks for discussion of the qualities of these tools and the posture and balance aspects of behaviour. Moreover, interpersonal communications lead to beneficial exchanges with the ethnographic experience. Although the American school of situated cognition (cognitive and psychological anthropology) is very useful to know for activity analysts, it should not be confused with EWA which, by definition, is directed towards an objective: knowing and transforming obstacles of all types that hinder and prevent satisfactory activities. First of all, EWA has to show, from the viewpoint of operators, how they build problems in order to be able to solve them. Ergonomists and ethnologists note how difficult this problem building may be in view of the variability of the technical system and of the state of operators' knowledge. The technology transfer situations studied by anthropotechnology need EWA even more in view of the frequent degradation of technical systems and the heterogeneous character of the two cultures present in the mind of the operator: his own culture and that which has inspired the imported technology. From this viewpoint, here and there the operator may be considered not as a performer but as the iterative creator of his task.     

Connectionism and the mind-body problem: exposing the distinction between mind and cognition

Does PDP-style connectionism imply that there are no such things as belief? Ramsey, Stich &; Garson (1991) have argued that it does, but their argument depends on a particular construal of the ontology of beliefs, according to which beliefs are discrete inner causal entities. In this paper I situate their argument within the broader issues of the clash of scientific and manifest images of the world, and the ontological mind-body problem. Their way of understanding belief places them in a long tradition of philosophers who conceive of mind and cognition as the same thing, a tradition that includes people as diverse as Descartes, Fodor, Churchland, and virtually all current cognitive scientists and cognitive neuroscientists. An alternative perspective on the nature of belief was championed by Ryle and is currently exemplified by Dennett's neo-instrumentalism. I argue that PDP-style connectionism, rather than implying that there are no beliefs, implies that the construal of the ontology of belief within the mind-as-cognition framework is mistaken. Beliefs are not discrete inner causal entities. Mind is not ontologically homogeneous; rather, like an economy, it is made up of a diversity of entities of very different ontological kinds interrelated in complex ways. Hence cognitive science is not the science of the mind; it is the science of cognition, which is only one aspect of mind.     

Construction and cross-correlation analysis of the affective physiological response database

Six kinds of emotions, namely joy, surprise, disgust, grief, anger and fear, are elicited from 300 ordinary college students by means of high-rated movie clips. The signals of galvanic skin response (GSR), heart rate (HR), blood volume pulse (BVP), electrocardiogram (ECG), respiration (Rsp.), facial electromyography (EMG) and two channels of electroencephalography (EEG) from the frontal lobe were recorded while the subjects were watching the affective eliciting materials. The samples of each kind of signals are analyzed by using random matrix theory (RMT). From the distribution of the eigenvalues and eigenvectors of the correlation matrices of BVP, facial EMG and two EEGs, we find that the statistical properties of these four kinds of signals coincide with the RMT predictions. However, the largest eigenvalue and the corresponding eigenvector component distribution of the correlation matrices of GSR, HR, ECG and Rsp. have deviated significantly from the RMT predictions. Our experiments also reveal that the correlations between two arbitrary samples of BVP, facial EMG or EEGs does not result from affective response but from random noises, whereas the amplitude variations of GSR, HR, ECG and Rsp. contain the correlated affective physiological response patterns. According to the data analysis results of RMT, 110 features are firstly extracted from the GSR and HR signals, and then a subset of the initial features is selected by using the Backward Selection algorithm. The affective data samples each of which is a vector of the features are classified by using a Fisher classifier, and furthermore, the user-independent emotion recognition systems with good prediction performance are constructed during the model selection process.     

情感生理反应样本库的建立与数据相关性分析

文中采用高收视率电影剪辑为唤起材料,以激发高兴、惊奇、厌恶、悲伤、愤怒和恐惧6种情感,记录被试观看情感唤起材料时的皮肤电导、心率、脉搏、心电、呼吸、面部肌电和额叶的两路脑电信号,获得了300名普通大学生的情感生理反应样本.用随机矩阵理论对各路信号数据进行相关性分析,发现皮肤电导、心率、心电和呼吸生理信号各自的相关矩阵最大特征值和对应特征向量的分布背离了随机矩阵理论的预测,表现出集体的两两互相关,揭示了由情感因素影响产生的相关生理反应;脉搏、面部肌电和额叶的两路脑电各自相关矩阵的特征值和特征向量没有明显背离随机矩阵理论预测,表明这4路信号各自数据之间的相关性由非情感因素的随机噪声引起,因此基于随机矩阵理论的数据分析为提取生理信号特征以研究情感的生理反应模式提供了可靠依据.研究中提取了皮肤电导和心率的110个初始特征,采用后向选择选出最能代表情感生理反应的信号特征子集,使用Fisher映射对情感生理反应样本进行分类,构造了具有较好预测性能的情感识别系统,获得了每一种情感较好区别于其他情感的生理信号特征组合.     

Communication Benchmarking and Performance Modelling of MPI Programs on Cluster Computers

This paper gives an overview of two related tools that we have developed to provide more accurate measurement and modelling of the performance of message-passing communication and application programs on distributed memory parallel computers. MPIBench uses a very precise, globally synchronised clock to measure the performance of MPI communication routines. It can generate probability distributions of communication times, not just the average values produced by other MPI benchmarks. This allows useful insights to be made into the MPI communication performance of parallel computers, and in particular how performance is affected by network contention. The Performance Evaluating Virtual Parallel Machine (PEVPM) provides a simple, fast and accurate technique for modelling and predicting the performance of message-passing parallel programs. It uses a virtual parallel machine to simulate the execution of the parallel program. The effects of network contention can be accurately modelled by sampling from the probability distributions generated by MPIBench. These tools are particularly useful on clusters with commodity Ethernet networks, where relatively high latencies, network congestion and TCP problems can significantly affect communication performance, which is difficult to model accurately using other tools. Experiments with example parallel programs demonstrate that PEVPM gives accurate performance predictions on commodity clusters. We also show that modelling communication performance using average times rather than sampling from probability distributions can give misleading results, particularly for programs running on a large number of processors.     

情感计算与工程理性*

通过对情感与理性的相容性分析以及情感在集成、适应、动力和交互等方面的功能研究,指出情感是人类解决现实复杂问题的有效机制。在情感功能与工程需求的适配性分析的基础上指出,围绕情感所展开的情感计算在复杂工程问题的求解上潜力巨大,具备有效性和一定的必要性。提出了情感计算在工程领域研究的本源性、相关性和可操作性策略,指出情感计算的工程实现需要对研究的切入点、情感发生机制、情感效应机制、平台选择等关键环节予以关注,并给出了相关建议。     

Beyond Tracking: Modelling Activity and Understanding Behaviour

In this work, we present a unified bottom-up and top-down automatic model selection based approach for modelling complex activities of multiple objects in cluttered scenes. An activity of multiple objects is represented based on discrete scene events and their behaviours are modelled by reasoning about the temporal and causal correlations among different events. This is significantly different from the majority of the existing techniques that are centred on object tracking followed by trajectory matching. In our approach, object-independent events are detected and classified by unsupervised clustering using Expectation-Maximisation (EM) and classified using automatic model selection based on Schwarz's Bayesian Information Criterion (BIC). Dynamic Probabilistic Networks (DPNs) are formulated for modelling the temporal and causal correlations among discrete events for robust and holistic scene-level behaviour interpretation. In particular, we developed a Dynamically Multi-Linked Hidden Markov Model (DML-HMM) based on the discovery of salient dynamic interlinks among multiple temporal processes corresponding to multiple event classes. A DML-HMM is built using BIC based factorisation resulting in its topology being intrinsically determined by the underlying causality and temporal order among events. Extensive experiments are conducted on modelling activities captured in different indoor and outdoor scenes. Our experimental results demonstrate that the performance of a DML-HMM on modelling group activities in a noisy and cluttered scene is superior compared to those of other comparable dynamic probabilistic networks including a Multi-Observation Hidden Markov Model (MOHMM), a Parallel Hidden Markov Model (PaHMM) and a Coupled Hidden Markov Model (CHMM). First online version published in February, 2006     

VeSense: High-performance and energy-efficient vehicular sensing platform

Although vehicular sensing where mobile users in vehicles continuously gather, process, and share location-sensitive and context-sensitive sensor data (e.g., street images, road condition, traffic flow) is emerging, little effort has been investigated in a model-based energy-efficient network paradigm of sensor information sharing in vehicular environments. Upon these optimization frameworks, a suite of optimization subproblems: a program partitioning and network resource allocation problem, we propose a distributed vehicular sensing platform, called VeSense where mobile users in vehicles publish/access sensor data via a cloud computing-based distributed P2P overlay network. The key objective is to satisfy the vehicular sensing application’s quality of service requirements by modeling each subsystem: mobile clients, wireless network medium, and distributed cloud services. By simulations based on experimental data, we present the proposed system can achieve up to 37 times more energy-efficient and 73 times faster compared to a standalone mobile application, in various vehicular sensing scenarios applying a realistic mobility model.     

Reality mining: sensing complex social systems

We introduce a system for sensing complex social systems with data collected from 100 mobile phones over the course of 9 months. We demonstrate the ability to use standard Bluetooth-enabled mobile telephones to measure information access and use in different contexts, recognize social patterns in daily user activity, infer relationships, identify socially significant locations, and model organizational rhythms.

面向交互式决策的情感计算与学习方法及其在过程控制工程中的应用

当今大量的工业过程控制工程问题, 如控制性能评价等的多目标决策问题通过人机交互得到解决. 然而, 在使用传统的交互式进化计算方法求解多目标决策问题时, 表现为局部搜索能力不强和过度依赖决策人员的主观意图. 本文提出一种新的情感计算方法和人机交互学习机制的解决方案. 提出一类基于刺激响应的情感计算模型(STAM), 并给出了情感空间和人的主观偏好之间的定量关系. 此外, 基于遗传算法框架, 建立情感交互学习策略, 旨在决策过程中逐步掌握人员的主观偏好, 降低人的疲劳程度, 使决策更加客观和科学. 附录A和B分别展示了情感学习算法的复杂度和收敛性分析. 为验证所提方法的正确性, 以测试函数及单回路反馈控制的PID参数整定问题进行研究, 得到了满意的结果, 验证了所提方法的可靠性与有效性.     

情感计算中的实验设计和情感度量方法研究

情感计算的目标是赋予计算机感知、理解和表达情感的能力,而情感数据的获取和情感真值的度量是情感计算的研究基础。为了更好地进行情感度量的研究,以FIFA足球电子游戏为场景,进行了精心设计的实验,即首先通过游戏场景诱发游戏者自然的情绪反应;接着提出了基于情绪脚本的情感编码方法,并结合范畴法和维度法对情感数据进行了自评和他评;最后对观察者间的一致性进行了统计分析。实验结果表明,该方法能够获得有效一致的情感度量结果。     

Incompleteness,Complexity, Randomness and Beyond

Gödel's Incompleteness Theorems have the same scientific status as Einstein's principle of relativity, Heisenberg's uncertainty principle, and Watson and Crick's double helix model of DNA. Our aim is to discuss some new faces of the incompleteness phenomenon unveiled by an information-theoretic approach to randomness and recent developments in quantum computing.