Learning Fine Motion by Markov Mixtures of Experts

Compliant control is a standard method for performing fine manipulation tasks, like grasping and assembly, but it requires estimation of the state of contact between the robot arm and the objects involved. Here we present a method to learn a model of the movement from measured data. The method requires little or no prior knowledge and the resulting model explicitly estimates the state of contact. The current state of contact is viewed as the hidden state variable of a discrete HMM. The control dependent transition probabilities between states are modeled as parametrized functions of the measurement We show that their parameters can be estimated from measurements concurrently with the estimation of the parameters of the movement in each state of contact. The learning algorithm is a variant of the EM procedure. The E step is computed exactly; solving the M step exactly would require solving a set of coupled nonlinear algebraic equations in the parameters. Instead, gradient ascent is used to produce an increase in likelihood.

Equivalence and Reduction of Hidden Markov Models

This report studies when and why two Hidden Markov Models (HMMs) may represent the same stochastic process. HMMs are characterized in terms of equivalence classes whose elements represent identical stochastic processes. This characterization yields polynomial time algorithms to detect equivalent HMMs. We also find fast algorithms to reduce HMMs to essentially unique and minimal canonical representations. The reduction to a canonical form leads to the definition of 'Generalized Markov Models' which are essentially HMMs without the positivity constraint on their parameters. We discuss how this generalization can yield more parsimonious representations of stochastic processes at the cost of the probabilistic interpretation of the model parameters.

Uso da geometria fractal na avaliação da variabilidade espacial de parâmetros de agregação do solo em São Paulo - Brasil.

O objetivo deste estudo foi explorar a aplicabilidade da teoria de fractais no estudo da variabilidade espacial em agregação de solo. A dimensão fractal D junto com RL que é o parâmetro que estima o tamanho do maior agregado foram usados como descritores de fragmentação. Estes valores estimados em diferentes locais na área experimental, foram interpolados usando krigagem ordinária e mapas de isolinhas construídos.

Enhancing the link between business and operations strategy.

This paper reprots on the use of banchmarking to improve the links between business and operations strategies. The use of benchmarking as a toll to facilitate improvements in these crucial links is examined. The existing literature on process benchmarking is used to from a structured questionnaire to apply to six case studies of major manuifacturing companies. Four of these case studies are presented in this paper to highlight the use of benchmarking in this application. Initial researh results are presented drawing upon the critical success factors indentified both in the literature and on the case results. Recommendations for further work are outlined

Benchmarking the business and operations strategy link.

This paper reports on the use of benchmarking to improve the links between business and operations strategies. The use of benchmarking as a tool to facilitate improvement in these crucial links is examined. The existing literature on process benchmarking is used to form a structured questionnaire to apply to six case studies of major maunfacturing companies. Four of these case studies are presented drawing upon the critical success factors identified both in the literature and on the case results. Recommendations for further work are outlined.

Fault identification through the combination of symbolic conflict recognition and Markov Chain-aided belief revision

F. Smith and Q. Shen. Fault identification through the combination of symbolic conflict recognition and Markov Chain-aided belief revision. IEEE Transactions on Systems, Man and Cybernetics, Part A: Systems and Humans, 34(5):649-663, 2004.

Semi-Supervised OWA Aggregation for Link-Based Similarity Evaluation and Alias Detection

T.Boongoen and Q. Shen. Semi-Supervised OWA Aggregation for Link-Based Similarity Evaluation and Alias Detection. Proceedings of the 18th International Conference on Fuzzy Systems (FUZZ-IEEE'09), pp. 288-293, 2009. Sponsorship: EPSRC

Vibrato in singing voice: the link between source-filter and sinusoidal models

The application of inverse filtering techniques for high-quality singing voice analysis/synthesis is discussed. In the context of source-filter models, inverse filtering provides a noninvasive method to extract the voice source, and thus to study voice quality. Although this approach is widely used in speech synthesis, this is not the case in singing voice. Several studies have proved that inverse filtering techniques fail in the case of singing voice, the reasons being unclear. In order to shed light on this problem, we will consider here an additional feature of singing voice, not present in speech: the vibrato. Vibrato has been traditionally studied by sinusoidal modeling. As an alternative, we will introduce here a novel noninteractive source filter model that incorporates the mechanisms of vibrato generation. This model will also allow the comparison of the results produced by inverse filtering techniques and by sinusoidal modeling, as they apply to singing voice and not to speech. In this way, the limitations of these conventional techniques, described in previous literature, will be explained. Both synthetic signals and singer recordings are used to validate and compare the techniques presented in the paper.

Alexander Mackay : missionary hero of Uganda.

http://www.archive.org/details/alexandermackay00unknuoft/

The life and work of Lucinda B. Helm : founder of the Woman's Parsonage and Home Mission Society of the M.E. church, South / by Arabel Wilbur Alexander.

http://purl.oclc.org/KUK/KDL/B92-53-27061877

Towards Eco-Dharma: The Contribution of Gandhian Thought to Ecological Ethics in India

Ecological concern prompts poor and indigenous people of India to consider how a society can ensure both protection of nature and their rightful claim for a just and sustainable future. Previous discussions defended the environment while ignoring the struggles of the poor for sustenance and their religious traditions and ethical values. Mohandas Karamchand Gandhi addressed similar socio-ecological concerns by adopting and adapting traditional religious and ethical notions to develop strategies for constructive, engaged resistance. The dissertation research and analysis verifies the continued relevance of the Gandhian understanding of dharma (ethics) in contemporary India as a basis for developing eco-dharma (eco-ethics) to link closely development, ecology, and religious values. The method of this study is interpretive, analytical, and critical. Françoise Houtart’s social analytical method is used to make visible and to suggest how to overcome social tensions from the perspective of marginalized and exploited peoples in India. The Indian government's development initiatives create a nexus between the eco-crisis and economic injustice, and communities’ responses. The Chipko movement seeks to protect the Himalayan forests from commercial logging. The Narmada Bachao Andolan strives to preserve the Narmada River and its forests and communities, where dam construction causes displacement. The use of Gandhian approaches by these movements provides a framework for integrating ecological concerns with people's struggles for survival. For Gandhi, dharma is a harmony of satya (truth), ahimsa (nonviolence), and sarvodaya (welfare of all). Eco-dharma is an integral, communitarian, and ecologically sensitive ethical paradigm. The study demonstrates that the Gandhian notion of dharma, implemented through nonviolent satyagraha (firmness in promoting truth), can direct community action that promotes responsible economic structures and the well-being of the biotic community and the environment. Eco-dharma calls for solidarity, constructive resistance, and ecologically and economically viable communities. The dissertation recommends that for a sustainable future, India must combine indigenous, appropriate, and small- or medium-scale industries as an alternative model of development in order to help reduce systemic poverty while enhancing ecological well-being.

Efficiently and Fairly Allocating Bandwidth at a Highly Congested Link

We consider the problem of efficiently and fairly allocating bandwidth at a highly congested link to a diverse set of flows, including TCP flows with various Round Trip Times (RTT), non-TCP-friendly flows such as Constant-Bit-Rate (CBR) applications using UDP, misbehaving, or malicious flows. Though simple, a FIFO queue management is vulnerable. Fair Queueing (FQ) can guarantee max-min fairness but fails at efficiency. RED-PD exploits the history of RED's actions in preferentially dropping packets from higher-rate flows. Thus, RED-PD attempts to achieve fairness at low cost. By relying on RED's actions, RED-PD turns out not to be effective in dealing with non-adaptive flows in settings with a highly heterogeneous mix of flows. In this paper, we propose a new approach we call RED-NB (RED with No Bias). RED-NB does not rely on RED's actions. Rather it explicitly maintains its own history for the few high-rate flows. RED-NB then adaptively adjusts flow dropping probabilities to achieve max-min fairness. In addition, RED-NB helps RED itself at very high loads by tuning RED's dropping behavior to the flow characteristics (restricted in this paper to RTTs) to eliminate its bias against long-RTT TCP flows while still taking advantage of RED's features at low loads. Through extensive simulations, we confirm the fairness of RED-NB and show that it outperforms RED, RED-PD, and CHOKe in all scenarios.