Bayesian online algorithms for learning in discrete Hidden Markov Models

We propose and analyze two different Bayesian online algorithms for learning in discrete Hidden Markov Models and compare their performance with the already known Baldi-Chauvin Algorithm. Using the Kullback-Leibler divergence as a measure of generalization we draw learning curves in simplified situations for these algorithms and compare their performances.

Online learning in discrete hidden Markov models

We present and analyze three different online algorithms for learning in discrete Hidden Markov Models (HMMs) and compare their performance with the Baldi-Chauvin Algorithm. Using the Kullback-Leibler divergence as a measure of the generalization error we draw learning curves in simplified situations and compare the results. The performance for learning drifting concepts of one of the presented algorithms is analyzed and compared with the Baldi-Chauvin algorithm in the same situations. A brief discussion about learning and symmetry breaking based on our results is also presented. © 2006 American Institute of Physics.

Analysing time series structure with hidden Markov models

This paper consides the problem of extracting the relationships between two time series in a non-linear non-stationary environment with Hidden Markov Models (HMMs). We describe an algorithm which is capable of identifying associations between variables. The method is applied both to synthetic data and real data. We show that HMMs are capable of modelling the oil drilling process and that they outperform existing methods.

Time delay estimation with hidden Markov models

Most traditional methods for extracting the relationships between two time series are based on cross-correlation. In a non-linear non-stationary environment, these techniques are not sufficient. We show in this paper how to use hidden Markov models to identify the lag (or delay) between different variables for such data. Adopting an information-theoretic approach, we develop a procedure for training HMMs to maximise the mutual information (MMI) between delayed time series. The method is used to model the oil drilling process. We show that cross-correlation gives no information and that the MMI approach outperforms maximum likelihood.

Hidden state models for time series

Amongst all the objectives in the study of time series, uncovering the dynamic law of its generation is probably the most important. When the underlying dynamics are not available, time series modelling consists of developing a model which best explains a sequence of observations. In this thesis, we consider hidden space models for analysing and describing time series. We first provide an introduction to the principal concepts of hidden state models and draw an analogy between hidden Markov models and state space models. Central ideas such as hidden state inference or parameter estimation are reviewed in detail. A key part of multivariate time series analysis is identifying the delay between different variables. We present a novel approach for time delay estimating in a non-stationary environment. The technique makes use of hidden Markov models and we demonstrate its application for estimating a crucial parameter in the oil industry. We then focus on hybrid models that we call dynamical local models. These models combine and generalise hidden Markov models and state space models. Probabilistic inference is unfortunately computationally intractable and we show how to make use of variational techniques for approximating the posterior distribution over the hidden state variables. Experimental simulations on synthetic and real-world data demonstrate the application of dynamical local models for segmenting a time series into regimes and providing predictive distributions.

Modelling data and voice traffic over IP networks using continuous-time Markov models

Common approaches to IP-traffic modelling have featured the use of stochastic models, based on the Markov property, which can be classified into black box and white box models based on the approach used for modelling traffic. White box models, are simple to understand, transparent and have a physical meaning attributed to each of the associated parameters. To exploit this key advantage, this thesis explores the use of simple classic continuous-time Markov models based on a white box approach, to model, not only the network traffic statistics but also the source behaviour with respect to the network and application. The thesis is divided into two parts: The first part focuses on the use of simple Markov and Semi-Markov traffic models, starting from the simplest two-state model moving upwards to n-state models with Poisson and non-Poisson statistics. The thesis then introduces the convenient to use, mathematically derived, Gaussian Markov models which are used to model the measured network IP traffic statistics. As one of the most significant contributions, the thesis establishes the significance of the second-order density statistics as it reveals that, in contrast to first-order density, they carry much more unique information on traffic sources and behaviour. The thesis then exploits the use of Gaussian Markov models to model these unique features and finally shows how the use of simple classic Markov models coupled with use of second-order density statistics provides an excellent tool for capturing maximum traffic detail, which in itself is the essence of good traffic modelling. The second part of the thesis, studies the ON-OFF characteristics of VoIP traffic with reference to accurate measurements of the ON and OFF periods, made from a large multi-lingual database of over 100 hours worth of VoIP call recordings. The impact of the language, prosodic structure and speech rate of the speaker on the statistics of the ON-OFF periods is analysed and relevant conclusions are presented. Finally, an ON-OFF VoIP source model with log-normal transitions is contributed as an ideal candidate to model VoIP traffic and the results of this model are compared with those of previously published work.

Delay estimation for multivariate time series

Most traditional methods for extracting the relationships between two time series are based on cross-correlation. In a non-linear non-stationary environment, these techniques are not sufficient. We show in this paper how to use hidden Markov models (HMMs) to identify the lag (or delay) between different variables for such data. We first present a method using maximum likelihood estimation and propose a simple algorithm which is capable of identifying associations between variables. We also adopt an information-theoretic approach and develop a novel procedure for training HMMs to maximise the mutual information between delayed time series. Both methods are successfully applied to real data. We model the oil drilling process with HMMs and estimate a crucial parameter, namely the lag for return.

Investigating the modulation of oral drug absorption using in vitro models

Dipeptides can be absorbed into cells via the dipeptide transporter (which also transported tripeptides and dipeptide derivatives). The optimum conditions for measuring the inhibition of Gly-Pro uptake in Caco-2 cells were identified. A number of structure-activity relationships were identified. These included the effects of increasing the amino-acid chain-length, and the presence of a thiol or hydroxyl group in the side-chain increased IC50 while the presence of a hydroxyl group did not. The benzyl esters had lower or equal IC50 values compared to the parent dipeptides while the methyl esters had higher values. These results indicated that while molecular properties did affect IC50, the size, charge and composition of three particular groups caused the most significant effects, supporting the structure-activity relationship identified. An assay was developed using calcein-AM to show the inhibition of p-glycoprotein activity. There was no significant change due to the presence of mannitol but there was in the presence of clyclosporin A (p<0.01). Incubating the cells with the test solution for 30 minutes before the addition of the ester resulted in a significant (p<0.001) difference. The assay was specific for p-glycoprotein, as the presence MRP inhibitors had no effect (p>0.05). The modified protocol allowed the identification of p-glycoprotein inhibitors quickly and simply using a cell suspension of unmodified cells. The clinically relevant buffering of grapefruit juice to pH 7 led to a four-fold increase in intracellular calcein and hence significant inhibition of p-glycoprotein. Buffered orange and lemon juices had no effect on the assay. Flavone derivatives had previously been found to be inhibitors of CYP3A4 yet neither naringin nor naringenin had any significant effect at concentrations found in grapefruit juice. Of the other (non-grapefruit) flavone derivatives tested, hesperidin, found in orange juice, had no significant effect, kaempferol and rutin also had no effect while genistein significantly inhibited p-glycoprotein (results that support previous studies). Hydroxycinnamic acids had no effect on p-glycoprotein. Studies on other compounds found that the balance between inhibiting p-glycoprotein and disrupting cell membranes depends on the compound containing an oxygen atom and the size of the negative charge on it, as well as three-dimensional arrangement of the atoms.

Wealth effects of mergers and acquisitions for US firms: using alternative pricing models

This empirical study employs a different methodology to examine the change in wealth associated with mergers and acquisitions (M&As) for US firms. Specifically, we employ the standard CAPM, the Fama-French three-factor model and the Carhart four-factor models within the OLS and GJR-GARCH estimation methods to test the behaviour of the cumulative abnormal returns (CARs). Whilst the standard CAPM captures the variability of stock returns with the overall market, the Fama-French factors capture the risk factors that are important to investors. Additionally, augmenting the Fama-French three-factor model with the Carhart momentum factor to generate the four-factor captures additional pricing elements that may affect stock returns. Traditionally, estimates of abnormal returns (ARs) in M&As situations rely on the standard OLS estimation method. However, the standard OLS will provide inefficient estimates of the ARs if the data contain ARCH and asymmetric effects. To minimise this problem of estimation efficiency we re-estimated the ARs using GJR-GARCH estimation method. We find that there is variation in the results both as regards the choice models and estimation methods. Besides these variations in the estimated models and the choice of estimation methods, we also tested whether the ARs are affected by the degree of liquidity of the stocks and the size of the firm. We document significant positive post-announcement cumulative ARs (CARs) for target firm shareholders under both the OLS and GJR-GARCH methods across all three methodologies. However, post-event CARs for acquiring firm shareholders were insignificant for both sets of estimation methods under the three methodologies. The GJR-GARCH method seems to generate larger CARs than those of the OLS method. Using both market capitalization and trading volume as a measure of liquidity and the size of the firm, we observed strong return continuations in the medium firms relative to small and large firms for target shareholders. We consistently observed market efficiency in small and large firm. This implies that target firms for small and large firms overreact to new information resulting in a more efficient market. For acquirer firms, our measure of liquidity captures strong return continuations for small firms under the OLS estimates for both CAPM and Fama-French three-factor models, whilst under the GJR-GARCH estimates only for Carhart model. Post-announcement bootstrapping simulated CARs confirmed our earlier results.

A hybrid generative/discriminative framework to train a semantic parser from an un-annotated corpus

We propose a hybrid generative/discriminative framework for semantic parsing which combines the hidden vector state (HVS) model and the hidden Markov support vector machines (HM-SVMs). The HVS model is an extension of the basic discrete Markov model in which context is encoded as a stack-oriented state vector. The HM-SVMs combine the advantages of the hidden Markov models and the support vector machines. By employing a modified K-means clustering method, a small set of most representative sentences can be automatically selected from an un-annotated corpus. These sentences together with their abstract annotations are used to train an HVS model which could be subsequently applied on the whole corpus to generate semantic parsing results. The most confident semantic parsing results are selected to generate a fully-annotated corpus which is used to train the HM-SVMs. The proposed framework has been tested on the DARPA Communicator Data. Experimental results show that an improvement over the baseline HVS parser has been observed using the hybrid framework. When compared with the HM-SVMs trained from the fully-annotated corpus, the hybrid framework gave a comparable performance with only a small set of lightly annotated sentences. © 2008. Licensed under the Creative Commons.

A novel framework of training hidden Markov support vector machines from lightly-annotated data

Natural language understanding (NLU) aims to map sentences to their semantic mean representations. Statistical approaches to NLU normally require fully-annotated training data where each sentence is paired with its word-level semantic annotations. In this paper, we propose a novel learning framework which trains the Hidden Markov Support Vector Machines (HM-SVMs) without the use of expensive fully-annotated data. In particular, our learning approach takes as input a training set of sentences labeled with abstract semantic annotations encoding underlying embedded structural relations and automatically induces derivation rules that map sentences to their semantic meaning representations. The proposed approach has been tested on the DARPA Communicator Data and achieved 93.18% in F-measure, which outperforms the previously proposed approaches of training the hidden vector state model or conditional random fields from unaligned data, with a relative error reduction rate of 43.3% and 10.6% being achieved.

Generalized methods and solvers for noise removal from piecewise constant signals. II. New methods

Removing noise from signals which are piecewise constant (PWC) is a challenging signal processing problem that arises in many practical scientific and engineering contexts. In the first paper (part I) of this series of two, we presented background theory building on results from the image processing community to show that the majority of these algorithms, and more proposed in the wider literature, are each associated with a special case of a generalized functional, that, when minimized, solves the PWC denoising problem. It shows how the minimizer can be obtained by a range of computational solver algorithms. In this second paper (part II), using this understanding developed in part I, we introduce several novel PWC denoising methods, which, for example, combine the global behaviour of mean shift clustering with the local smoothing of total variation diffusion, and show example solver algorithms for these new methods. Comparisons between these methods are performed on synthetic and real signals, revealing that our new methods have a useful role to play. Finally, overlaps between the generalized methods of these two papers and others such as wavelet shrinkage, hidden Markov models, and piecewise smooth filtering are touched on.

Rotation invariant iris feature extraction using Gaussian Markov random fields with non-separable wavelet

Rotation invariance is important for an iris recognition system since changes of head orientation and binocular vergence may cause eye rotation. The conventional methods of iris recognition cannot achieve true rotation invariance. They only achieve approximate rotation invariance by rotating the feature vector before matching or unwrapping the iris ring at different initial angles. In these methods, the complexity of the method is increased, and when the rotation scale is beyond the certain scope, the error rates of these methods may substantially increase. In order to solve this problem, a new rotation invariant approach for iris feature extraction based on the non-separable wavelet is proposed in this paper. Firstly, a bank of non-separable orthogonal wavelet filters is used to capture characteristics of the iris. Secondly, a method of Markov random fields is used to capture rotation invariant iris feature. Finally, two-class kernel Fisher classifiers are adopted for classification. Experimental results on public iris databases show that the proposed approach has a low error rate and achieves true rotation invariance. © 2010.

Variational mean-field algorithm for efficient inference in large systems of stochastic differential equations

This work introduces a Gaussian variational mean-field approximation for inference in dynamical systems which can be modeled by ordinary stochastic differential equations. This new approach allows one to express the variational free energy as a functional of the marginal moments of the approximating Gaussian process. A restriction of the moment equations to piecewise polynomial functions, over time, dramatically reduces the complexity of approximate inference for stochastic differential equation models and makes it comparable to that of discrete time hidden Markov models. The algorithm is demonstrated on state and parameter estimation for nonlinear problems with up to 1000 dimensional state vectors and compares the results empirically with various well-known inference methodologies.

Bandwidth management of WiMAX systems and performance modeling

WiMAX has been introduced as a competitive alternative for metropolitan broadband wireless access technologies. It is connection oriented and it can provide very high data rates, large service coverage, and flexible quality of services (QoS). Due to the large number of connections and flexible QoS supported by WiMAX, the uplink access in WiMAX networks is very challenging since the medium access control (MAC) protocol must efficiently manage the bandwidth and related channel allocations. In this paper, we propose and investigate a cost-effective WiMAX bandwidth management scheme, named the WiMAX partial sharing scheme (WPSS), in order to provide good QoS while achieving better bandwidth utilization and network throughput. The proposed bandwidth management scheme is compared with a simple but inefficient scheme, named the WiMAX complete sharing scheme (WCPS). A maximum entropy (ME) based analytical model (MEAM) is proposed for the performance evaluation of the two bandwidth management schemes. The reason for using MEAM for the performance evaluation is that MEAM can efficiently model a large-scale system in which the number of stations or connections is generally very high, while the traditional simulation and analytical (e.g., Markov models) approaches cannot perform well due to the high computation complexity. We model the bandwidth management scheme as a queuing network model (QNM) that consists of interacting multiclass queues for different service classes. Closed form expressions for the state and blocking probability distributions are derived for those schemes. Simulation results verify the MEAM numerical results and show that WPSS can significantly improve the network's performance compared to WCPS.