5 resultados para Probability Density Function
em Universidade Federal do Rio Grande do Norte(UFRN)
Resumo:
In this work, we propose a two-stage algorithm for real-time fault detection and identification of industrial plants. Our proposal is based on the analysis of selected features using recursive density estimation and a new evolving classifier algorithm. More specifically, the proposed approach for the detection stage is based on the concept of density in the data space, which is not the same as probability density function, but is a very useful measure for abnormality/outliers detection. This density can be expressed by a Cauchy function and can be calculated recursively, which makes it memory and computational power efficient and, therefore, suitable for on-line applications. The identification/diagnosis stage is based on a self-developing (evolving) fuzzy rule-based classifier system proposed in this work, called AutoClass. An important property of AutoClass is that it can start learning from scratch". Not only do the fuzzy rules not need to be prespecified, but neither do the number of classes for AutoClass (the number may grow, with new class labels being added by the on-line learning process), in a fully unsupervised manner. In the event that an initial rule base exists, AutoClass can evolve/develop it further based on the newly arrived faulty state data. In order to validate our proposal, we present experimental results from a level control didactic process, where control and error signals are used as features for the fault detection and identification systems, but the approach is generic and the number of features can be significant due to the computationally lean methodology, since covariance or more complex calculations, as well as storage of old data, are not required. The obtained results are significantly better than the traditional approaches used for comparison
Resumo:
In this work we study the Hidden Markov Models with finite as well as general state space. In the finite case, the forward and backward algorithms are considered and the probability of a given observed sequence is computed. Next, we use the EM algorithm to estimate the model parameters. In the general case, the kernel estimators are used and to built a sequence of estimators that converge in L1-norm to the density function of the observable process
Resumo:
In this work we studied the asymptotic unbiasedness, the strong and the uniform strong consistencies of a class of kernel estimators fn as an estimator of the density function f taking values on a k-dimensional sphere
Resumo:
This present work uses a generalized similarity measure called correntropy to develop a new method to estimate a linear relation between variables given their samples. Towards this goal, the concept of correntropy is extended from two variables to any two vectors (even with different dimensions) using a statistical framework. With this multidimensionals extensions of Correntropy the regression problem can be formulated in a different manner by seeking the hyperplane that has maximum probability density with the target data. Experiments show that the new algorithm has a nice fixed point update for the parameters and robust performs in the presence of outlier noise.
Resumo:
Nicotine administration in humans and rodents enhances memory and attention, and also has a positive effect in Alzheimer's Disease. The Medial Septum / Diagonal Band of Broca complex (MS/DBB) – a main cholinergic system – massively projects to the hippocampus through the fimbria-fornix, and this pathway is called the septohippocampal pathway. It has been demonstrated that the MS/DBB acts directly on the local field potential (LFP) rhythmic organization of the hippocampus, especially in the rhythmogenesis of Theta (4-8Hz) – an oscillation intrinsically linked to hippocampus mnemonic function. In vitro experiments gave evidence that nicotine applied to the MS/DBB generates a local network Theta rhythm within the MS/DBB. Thus, the present study proposes to elucidate the function of nicotine in the MS/DBB on the septo-hippocampal pathway. In vivo experiments compared the effect of MS/DBB microinfusion of saline (n=5) and nicotine (n=8) on Ketamine/Xylazine anaesthetized mice. We observed power spectrum density in the Gamma range (35 to 55 Hz) increasing in both structures (Wilcoxon Rank-Sum test, p=0.038) but with no change in coherence between these structures in the same range (Wilcoxon Rank-Sum test, p=0.60). There was also a decrease in power of the ketamineinduced Delta oscillation (1 to 3 Hz). We also performed in vitro experiments on the effect of nicotine on membrane voltage and action potential. We patch-clamped 22 neurons in current-clamp mode; 12 neurons were responsive to nicotine, half of them increased firing rate and other 6 decreased, and they significantly differed in action potential threshold (-47.3±0.9 mV vs. -41±1.9 mV, respectively, p=0.007) and halfwidth time (1.6±0.08 ms vs. 2±0.12 ms, respectively, p=0.01). Furthermore, we performed another set of in vitro experiments concerning the connectivity of the three major neuronal populations of MS/DBB that use acetylcholine, GABA or glutamate as neurotransmitter. Paired patch-clamp recordings found that glutamatergic and GABAergic neurons realize intra-septal connections that produce sizable currents in MS/DBB postsynaptic neurons. The probability of connectivity between different neuronal populations gave rise to a MS/DBB topology that was implemented in a realistic model, which corroborates that the network is highly sensitive to the generation of Gamma rhythm. Together, the data available in the full set of experiments suggests that nicotine may act as a cognitive enhancer, by inducing gamma oscillation in the local circuitry of the MS/DBB.