Clustering Based Large Margin Classification: A Scalable Approach using SOCP Formulation

This paper presents a novel Second Order Cone Programming (SOCP) formulation for large scale binary classification tasks. Assuming that the class conditional densities are mixture distributions, where each component of the mixture has a spherical covariance, the second order statistics of the components can be estimated efficiently using clustering algorithms like BIRCH. For each cluster, the second order moments are used to derive a second order cone constraint via a Chebyshev-Cantelli inequality. This constraint ensures that any data point in the cluster is classified correctly with a high probability. This leads to a large margin SOCP formulation whose size depends on the number of clusters rather than the number of training data points. Hence, the proposed formulation scales well for large datasets when compared to the state-of-the-art classifiers, Support Vector Machines (SVMs). Experiments on real world and synthetic datasets show that the proposed algorithm outperforms SVM solvers in terms of training time and achieves similar accuracies.

Fully complex-valued ELM classifiers for human action recognition

In this paper, we present a fast learning neural network classifier for human action recognition. The proposed classifier is a fully complex-valued neural network with a single hidden layer. The neurons in the hidden layer employ the fully complex-valued hyperbolic secant as an activation function. The parameters of the hidden layer are chosen randomly and the output weights are estimated analytically as a minimum norm least square solution to a set of linear equations. The fast leaning fully complex-valued neural classifier is used for recognizing human actions accurately. Optical flow-based features extracted from the video sequences are utilized to recognize 10 different human actions. The feature vectors are computationally simple first order statistics of the optical flow vectors, obtained from coarse to fine rectangular patches centered around the object. The results indicate the superior performance of the complex-valued neural classifier for action recognition. The superior performance of the complex neural network for action recognition stems from the fact that motion, by nature, consists of two components, one along each of the axes.

Robust formulations for clustering-based large-scale classification

Chebyshev-inequality-based convex relaxations of Chance-Constrained Programs (CCPs) are shown to be useful for learning classifiers on massive datasets. In particular, an algorithm that integrates efficient clustering procedures and CCP approaches for computing classifiers on large datasets is proposed. The key idea is to identify high density regions or clusters from individual class conditional densities and then use a CCP formulation to learn a classifier on the clusters. The CCP formulation ensures that most of the data points in a cluster are correctly classified by employing a Chebyshev-inequality-based convex relaxation. This relaxation is heavily dependent on the second-order statistics. However, this formulation and in general such relaxations that depend on the second-order moments are susceptible to moment estimation errors. One of the contributions of the paper is to propose several formulations that are robust to such errors. In particular a generic way of making such formulations robust to moment estimation errors is illustrated using two novel confidence sets. An important contribution is to show that when either of the confidence sets is employed, for the special case of a spherical normal distribution of clusters, the robust variant of the formulation can be posed as a second-order cone program. Empirical results show that the robust formulations achieve accuracies comparable to that with true moments, even when moment estimates are erroneous. Results also illustrate the benefits of employing the proposed methodology for robust classification of large-scale datasets.

Eigen-profiles of spatio-temporal fragments for adaptive region-based tracking

We propose a novel space-time descriptor for region-based tracking which is very concise and efficient. The regions represented by covariance matrices within a temporal fragment, are used to estimate this space-time descriptor which we call the Eigenprofiles(EP). EP so obtained is used in estimating the Covariance Matrix of features over spatio-temporal fragments. The Second Order Statistics of spatio-temporal fragments form our target model which can be adapted for variations across the video. The model being concise also allows the use of multiple spatially overlapping fragments to represent the target. We demonstrate good tracking results on very challenging datasets, shot under insufficient illumination conditions.

Conductivity noise study of the insulator-metal transition and phase coexistence in epitaxial samarium nickelate thin films

Interaction between the lattice and the orbital degrees of freedom not only makes rare-earth nickelates unusually ``bad metal,'' but also introduces a temperature-driven insulator-metal phase transition. Here we investigate this insulator-metal phase transition in thin films of SmNiO3 using the slow time-dependent fluctuations (noise) in resistivity. The normalized magnitude of noise is found to be extremely large, being nearly eight orders of magnitude higher than thin films of common disordered metallic systems, and indicates electrical conduction via classical percolation in a spatially inhomogeneous medium. The higher-order statistics of the fluctuations indicate a strong non-Gaussian component of noise close to the transition, attributing the inhomogeneity to the coexistence of the metallic and insulating phases. Our experiment offers insight into the impact of lattice-orbital coupling on the microscopic mechanism of electron transport in the rare-earth nickelates.

Large-eddy simulations of fully developed turbulent channel and pipe flows with smooth and rough walls

Studies in turbulence often focus on two flow conditions, both of which occur frequently in real-world flows and are sought-after for their value in advancing turbulence theory. These are the high Reynolds number regime and the effect of wall surface roughness. In this dissertation, a Large-Eddy Simulation (LES) recreates both conditions over a wide range of Reynolds numbers Re_τ = O(10²)-O(10⁸) and accounts for roughness by locally modeling the statistical effects of near-wall anisotropic fine scales in a thin layer immediately above the rough surface. A subgrid, roughness-corrected wall model is introduced to dynamically transmit this modeled information from the wall to the outer LES, which uses a stretched-vortex subgrid-scale model operating in the bulk of the flow. Of primary interest is the Reynolds number and roughness dependence of these flows in terms of first and second order statistics. The LES is first applied to a fully turbulent uniformly-smooth/rough channel flow to capture the flow dynamics over smooth, transitionally rough and fully rough regimes. Results include a Moody-like diagram for the wall averaged friction factor, believed to be the first of its kind obtained from LES. Confirmation is found for experimentally observed logarithmic behavior in the normalized stream-wise turbulent intensities. Tight logarithmic collapse, scaled on the wall friction velocity, is found for smooth-wall flows when Re_τ ≥ O(10⁶) and in fully rough cases. Since the wall model operates locally and dynamically, the framework is used to investigate non-uniform roughness distribution cases in a channel, where the flow adjustments to sudden surface changes are investigated. Recovery of mean quantities and turbulent statistics after transitions are discussed qualitatively and quantitatively at various roughness and Reynolds number levels. The internal boundary layer, which is defined as the border between the flow affected by the new surface condition and the unaffected part, is computed, and a collapse of the profiles on a length scale containing the logarithm of friction Reynolds number is presented. Finally, we turn to the possibility of expanding the present framework to accommodate more general geometries. As a first step, the whole LES framework is modified for use in the curvilinear geometry of a fully-developed turbulent pipe flow, with implementation carried out in a spectral element solver capable of handling complex wall profiles. The friction factors have shown favorable agreement with the superpipe data, and the LES estimates of the Karman constant and additive constant of the log-law closely match values obtained from experiment.

Optimum linear coding for additive noise systems using information feedback

The feedback coding problem for Gaussian systems in which the noise is neither white nor statistically independent between channels is formulated in terms of arbitrary linear codes at the transmitter and at the receiver. This new formulation is used to determine a number of feedback communication systems. In particular, the optimum linear code that satisfies an average power constraint on the transmitted signals is derived for a system with noiseless feedback and forward noise of arbitrary covariance. The noisy feedback problem is considered and signal sets for the forward and feedback channels are obtained with an average power constraint on each. The general formulation and results are valid for non-Gaussian systems in which the second order statistics are known, the results being applicable to the determination of error bounds via the Chebychev inequality.

Separação cega de sinais de fala utilizando detectores de voz.

Neste trabalho contemplamos o emprego de detectores de voz como uma etapa de pré- processamento de uma técnica de separação cega de sinais implementada no domínio do tempo, que emprega estatísticas de segunda ordem para a separação de misturas convolutivas e determinadas. Seu algoritmo foi adaptado para realizar a separação tanto em banda cheia quanto em sub-bandas, considerando a presença e a ausência de instantes de silêncio em misturas de sinais de voz. A ideia principal consiste em detectar trechos das misturas que contenham atividade de voz, evitando que o algoritmo de separação seja acionado na ausência de voz, promovendo ganho de desempenho e redução do custo computacional.

Measurement-Integrated simulations and Kalman filter applied to a co-flowing jet

This paper deals with the experimental evaluation of a flow analysis system based on the integration between an under-resolved Navier-Stokes simulation and experimental measurements with the mechanism of feedback (referred to as Measurement-Integrated simulation), applied to the case of a planar turbulent co-flowing jet. The experiments are performed with inner-to-outer-jet velocity ratio around 2 and the Reynolds number based on the inner-jet heights about 10000. The measurement system is a high-speed PIV, which provides time-resolved data of the flow-field, on a field of view which extends to 20 jet heights downstream the jet outlet. The experimental data can thus be used both for providing the feedback data for the simulations and for validation of the MI-simulations over a wide region. The effect of reduced data-rate and spatial extent of the feedback (i.e. measurements are not available at each simulation time-step or discretization point) was investigated. At first simulations were run with full information in order to obtain an upper limit of the MI-simulations performance. The results show the potential of this methodology of reproducing first and second order statistics of the turbulent flow with good accuracy. Then, to deal with the reduced data different feedback strategies were tested. It was found that for small data-rate reduction the results are basically equivalent to the case of full-information feedback but as the feedback data-rate is reduced further the error increases and tend to be localized in regions of high turbulent activity. Moreover, it is found that the spatial distribution of the error looks qualitatively different for different feedback strategies. Feedback gain distributions calculated by optimal control theory are presented and proposed as a mean to make it possible to perform MI-simulations based on localized measurements only. So far, we have not been able to low error between measurements and simulations by using these gain distributions.

2D PIV measurements in the near field of grid turbulence using stitched fields from multiple cameras

We present measurements of grid turbulence using 2D particle image velocimetry taken immediately downstream from the grid at a Reynolds number of Re M = 16500 where M is the rod spacing. A long field of view of 14M x 4M in the down- and cross-stream directions was achieved by stitching multiple cameras together. Two uniform biplanar grids were selected to have the same M and pressure drop but different rod diameter D and crosssection. A large data set (10 4 vector fields) was obtained to ensure good convergence of second-order statistics. Estimations of the dissipation rate ε of turbulent kinetic energy (TKE) were found to be sensitive to the number of meansquared velocity gradient terms included and not whether the turbulence was assumed to adhere to isotropy or axisymmetry. The resolution dependency of different turbulence statistics was assessed with a procedure that does not rely on the dissipation scale η. The streamwise evolution of the TKE components and ε was found to collapse across grids when the rod diameter was included in the normalisation. We argue that this should be the case between all regular grids when the other relevant dimensionless quantities are matched and the flow has become homogeneous across the stream. Two-point space correlation functions at x/M = 1 show evidence of complex wake interactions which exhibit a strong Reynolds number dependence. However, these changes in initial conditions disappear indicating rapid cross-stream homogenisation. On the other hand, isotropy was, as expected, not found to be established by x/M = 12 for any case studied. © Springer-Verlag 2012.

An Information Theoretic Framework for Field Monitoring Using Autonomously Mobile Sensors

We consider a mobile sensor network monitoring a spatio-temporal field. Given limited cache sizes at the sensor nodes, the goal is to develop a distributed cache management algorithm to efficiently answer queries with a known probability distribution over the spatial dimension. First, we propose a novel distributed information theoretic approach in which the nodes locally update their caches based on full knowledge of the space-time distribution of the monitored phenomenon. At each time instant, local decisions are made at the mobile nodes concerning which samples to keep and whether or not a new sample should be acquired at the current location. These decisions account for minimizing an entropic utility function that captures the average amount of uncertainty in queries given the probability distribution of query locations. Second, we propose a different correlation-based technique, which only requires knowledge of the second-order statistics, thus relaxing the stringent constraint of having a priori knowledge of the query distribution, while significantly reducing the computational overhead. It is shown that the proposed approaches considerably improve the average field estimation error by maintaining efficient cache content. It is further shown that the correlation-based technique is robust to model mismatch in case of imperfect knowledge of the underlying generative correlation structure.

Modeling the Possible Influences of Eye Movements on the Refinement of Cortical Direction Selectivity

The second-order statistics of neural activity was examined in a model of the cat LGN and V1 during free-viewing of natural images. In the model, the specific patterns of thalamocortical activity required for a Bebbian maturation of direction-selective cells in VI were found during the periods of visual fixation, when small eye movements occurred, but not when natural images were examined in the absence of fixational eye movements. In addition, simulations of stroboscopic reming that replicated the abnormal pattern of eye movements observed in kittens chronically exposed to stroboscopic illumination produced results consistent with the reported loss of direction selectivity and preservation of orientation selectivity. These results suggest the involvement of the oculomotor activity of visual fixation in the maturation of cortical direction selectivity.

Investment funds: What next?

Investment funds provide a low cost method of sharing in the rewards from capitalism. Recently “alternative investments” such as hedge funds have grown rapidly and the trading strategies open to hedge funds are now becoming available to mutual funds and even to ordinary retail investors. In this paper we analyze problems in assessing fund performance and the prospects for investment fund sectors. Choosing genuine outperformers among top funds requires a careful assessment of non-normality, order statistics and the possibility of false discoveries. The risk adjusted performance of the average hedge fund over the last 10-15 is actually not that impressive, although the “top” funds do appear to have statistically significant positive alphas.