A Bacterial Foraging Optimization and Learning Automata Based Feature Selection for Motor Imagery EEG Classification

Selection of relevant features is an open problem in Brain-computer interfacing (BCI) research. Sometimes, features extracted from brain signals are high dimensional which in turn affects the accuracy of the classifier. Selection of the most relevant features improves the performance of the classifier and reduces the computational cost of the system. In this study, we have used a combination of Bacterial Foraging Optimization and Learning Automata to determine the best subset of features from a given motor imagery electroencephalography (EEG) based BCI dataset. Here, we have employed Discrete Wavelet Transform to obtain a high dimensional feature set and classified it by Distance Likelihood Ratio Test. Our proposed feature selector produced an accuracy of 80.291% in 216 seconds.

SPEECH ENHANCEMENT USING DISCRIMINATIVE RANDOM FIELDS

Speech enhancement in stationary noise is addressed using the ideal channel selection framework. In order to estimate the binary mask, we propose to classify each time-frequency (T-F) bin of the noisy signal as speech or noise using Discriminative Random Fields (DRF). The DRF function contains two terms - an enhancement function and a smoothing term. On each T-F bin, we propose to use an enhancement function based on likelihood ratio test for speech presence, while Ising model is used as smoothing function for spectro-temporal continuity in the estimated binary mask. The effect of the smoothing function over successive iterations is found to reduce musical noise as opposed to using only enhancement function. The binary mask is inferred from the noisy signal using Iterated Conditional Modes (ICM) algorithm. Sentences from NOIZEUS corpus are evaluated from 0 dB to 15 dB Signal to Noise Ratio (SNR) in 4 kinds of additive noise settings: additive white Gaussian noise, car noise, street noise and pink noise. The reconstructed speech using the proposed technique is evaluated in terms of average segmental SNR, Perceptual Evaluation of Speech Quality (PESQ) and Mean opinion Score (MOS).

Fiabilidad de los predictores clínicos y de la biopsia de arteria temporal en el diagnóstico de la arteritis de células gigantes

182 p. : il.

Essays in Behavioral Decision Theory

This thesis studies decision making under uncertainty and how economic agents respond to information. The classic model of subjective expected utility and Bayesian updating is often at odds with empirical and experimental results; people exhibit systematic biases in information processing and often exhibit aversion to ambiguity. The aim of this work is to develop simple models that capture observed biases and study their economic implications.

In the first chapter I present an axiomatic model of cognitive dissonance, in which an agent's response to information explicitly depends upon past actions. I introduce novel behavioral axioms and derive a representation in which beliefs are directionally updated. The agent twists the information and overweights states in which his past actions provide a higher payoff. I then characterize two special cases of the representation. In the first case, the agent distorts the likelihood ratio of two states by a function of the utility values of the previous action in those states. In the second case, the agent's posterior beliefs are a convex combination of the Bayesian belief and the one which maximizes the conditional value of the previous action. Within the second case a unique parameter captures the agent's sensitivity to dissonance, and I characterize a way to compare sensitivity to dissonance between individuals. Lastly, I develop several simple applications and show that cognitive dissonance contributes to the equity premium and price volatility, asymmetric reaction to news, and belief polarization.

The second chapter characterizes a decision maker with sticky beliefs. That is, a decision maker who does not update enough in response to information, where enough means as a Bayesian decision maker would. This chapter provides axiomatic foundations for sticky beliefs by weakening the standard axioms of dynamic consistency and consequentialism. I derive a representation in which updated beliefs are a convex combination of the prior and the Bayesian posterior. A unique parameter captures the weight on the prior and is interpreted as the agent's measure of belief stickiness or conservatism bias. This parameter is endogenously identified from preferences and is easily elicited from experimental data.

The third chapter deals with updating in the face of ambiguity, using the framework of Gilboa and Schmeidler. There is no consensus on the correct way way to update a set of priors. Current methods either do not allow a decision maker to make an inference about her priors or require an extreme level of inference. In this chapter I propose and axiomatize a general model of updating a set of priors. A decision maker who updates her beliefs in accordance with the model can be thought of as one that chooses a threshold that is used to determine whether a prior is plausible, given some observation. She retains the plausible priors and applies Bayes' rule. This model includes generalized Bayesian updating and maximum likelihood updating as special cases.

Integrating methods for determining length-at-age to improve growth estimates for two large scombrids

Fish growth is commonly estimated from length-at-age data obtained from otoliths. There are several techniques for estimating length-at-age from otoliths including 1) direct observed counts of annual increments; 2) age adjustment based on a categorization of otolith margins; 3) age adjustment based on known periods of spawning and annuli formation; 4) back-calculation to all annuli, and 5) back-calculation to the last annulus only. In this study we compared growth estimates (von Bertalanffy growth functions) obtained from the above five methods for estimating length-at-age from otoliths for two large scombrids: narrow-barred Spanish mackerel (Scomberomorus commerson) and broad-barred king mackerel (Scomberomorus semifasciatus). Likelihood ratio tests revealed that the largest differences in growth occurred between the back-calculation methods and the observed and adjusted methods for both species of mackerel. The pattern, however, was more pronounced for S. commerson than for S. semifasciatus, because of the pronounced effect of gear selectivity demonstrated for S. commerson. We propose a method of substituting length-at-age data from observed or adjusted methods with back-calculated length-at-age data to provide more appropriate estimates of population growth than those obtained with the individual methods alone, particularly when faster growing young fish are disproportionately selected for. Substitution of observed or adjusted length-at-age data with back-calculated length-at-age data provided more realistic estimates of length for younger ages than observed or adjusted methods as well as more realistic estimates of mean maximum length than those derived from backcalculation methods alone.

Estimates of growth and comparisons of growth rates determined from length- and age-based models for populations of purple wrasse (Notolabrus fucicola)

Growth of a temperate reefa-ssociated fish, the purple wrasse (Notolabrus fucicola), was examined from two sites on the east coast of Tasmania by using age- and length-based models. Models based on the von Bertalanffy growth function, in the standard and a reparameterized form, were constructed by using otolith-derived age estimates. Growth trajectories from tag-recaptures were used to construct length-based growth models derived from the GROTAG model, in turn a reparameterization of the Fabens model. Likelihood ratio tests (LRTs) determined the optimal parameterization of the GROTAG model, including estimators of individual growth variability, seasonal growth, measurement error, and outliers for each data set. Growth models and parameter estimates were compared by bootstrap confidence intervals, LRTs, and randomization tests and plots of bootstrap parameter estimates. The relative merit of these methods for comparing models and parameters was evaluated; LRTs combined with bootstrapping and randomization tests provided the most insight into the relationships between parameter estimates. Significant differences in growth of purple wrasse were found between sites in both length- and age-based models. A significant difference in the peak growth season was found between sites, and a large difference in growth rate between sexes was found at one site with the use of length-based models.

Reproductive biology of carpenter seabream (Argyrozona argyrozona) (Pisces: Sparidae) in a marine protected area

The carpenter seabream (Argyrozona argyrozona) is an endemic South African sparid that comprises an important part of the handline fishery. A three-year study (1998−2000) into its reproductive biology within the Tsitsikamma National Park revealed that these fishes are serial spawning late gonochorists. The size at 50% maturity (L50) was estimated at 292 and 297 mm FL for both females and males, respectively. A likelihood ratio test revealed that there was no significant difference between male and female L50 (P>0.5). Both monthly gonadosomatic indices and macroscopically determined ovarian stages strongly indicate that A. argyrozona within the Tsitsikamma National Park spawn in the astral summer between November and April. The presence of postovulatory follicles (POFs) confirmed a six-month spawning season, and monthly proportions of early (0−6 hour old) POFs showed that spawning frequency was highest (once every 1−2 days) from December to March. Although spawning season was more highly correlated to photoperiod (r = 0.859) than temperature (r = −0.161), the daily proportion of spawning fish was strongly correlated (r= 0.93) to ambient temperature over the range 9−22oC. These results indicate that short-term upwelling events, a strong feature in the Tsitsikamma National Park during summer, may negatively affect carpenter fecundity. Both spawning frequency and duration (i.e., length of spawning season) increased with fish length. As a result of the allometric relationship between annual fecundity and fish mass a 3-kg fish was calculated to produce fivefold more eggs per kilogram of body weight than a fish of 1 kg. In addition to producing more eggs per unit of weight each year, larger fish also produce significantly larger eggs.

A general framework for integrating environmental time series into stock assessment models: model description, simulation testing, and example

We present a method to integrate environmental time series into stock assessment models and to test the significance of correlations between population processes and the environmental time series. Parameters that relate the environmental time series to population processes are included in the stock assessment model, and likelihood ratio tests are used to determine if the parameters improve the fit to the data significantly. Two approaches are considered to integrate the environmental relationship. In the environmental model, the population dynamics process (e.g. recruitment) is proportional to the environmental variable, whereas in the environmental model with process error it is proportional to the environmental variable, but the model allows an additional temporal variation (process error) constrained by a log-normal distribution. The methods are tested by using simulation analysis and compared to the traditional method of correlating model estimates with environmental variables outside the estimation procedure. In the traditional method, the estimates of recruitment were provided by a model that allowed the recruitment only to have a temporal variation constrained by a log-normal distribution. We illustrate the methods by applying them to test the statistical significance of the correlation between sea-surface temperature (SST) and recruitment to the snapper (Pagrus auratus) stock in the Hauraki Gulf–Bay of Plenty, New Zealand. Simulation analyses indicated that the integrated approach with additional process error is superior to the traditional method of correlating model estimates with environmental variables outside the estimation procedure. The results suggest that, for the snapper stock, recruitment is positively correlated with SST at the time of spawning.

Seasonal growth of King George whiting (Sillaginodes punctata) estimated from length-at-age samples of the legal-size harvest

Samples of 11,000 King George whiting (Sillaginodes punctata) from the South Australian commercial and recreational catch, supplemented by research samples, were aged from otoliths. Samples were analyzed from three coastal regions and by sex. Most sampling was undertaken at fish processing plants, from which only fish longer than the legal minimum length were obtained. A left-truncated normal distribution of lengths at monthly age was therefore employed as model likelihood. Mean length-at-monthly-age was described by a generalized von Bertalanffy formula with sinusoidal seasonality. Likelihood standard deviation was modeled to vary allometrically with mean length. A range of related formulas (with 6 to 8 parameters) for seasonal mean length at age were compared. In addition to likelihood ratio tests of relative fit, model selection criteria were a minimum occurrence of high uncertainties (>20% SE), of high correlations (>0.9, >0.95, and >0.99) and of parameter estimates at their biological limits, and we sought a model with a minimum number of parameters. A generalized von Bertalanffy formula with t0 fixed at 0 was chosen. The truncated likelihood alleviated the overestimation bias of mean length at age that would otherwise accrue from catch samples being restricted to legal sizes.

Comparative molecular population genetics of phycoerythrin locus in Prochlorococcus

As the only remainder type of phycobiliproteins in Prochlorococcus, the actual role of phycoerythrin still remains unknown. Previous studies revealed that two different forms of phycoerythrin gene were found in two ecotypes of Prochlorococcus that are specifically adapted to either high light (HL) or low light (LL) conditions. Here we analyze patterns of phycoerythrin nucleotide variation in the HL- and LL-Prochlorococcus populations. Our analyses reveal a significantly greater number of non-synonymous fixed substitutions in peB and peA than expected based on interspecific comparisons. This pattern of excess non-synonymous fixed substitutions is not seen in other five phycoerythrin-related genes (peZ/V/Y/T/S). Several neutrality statistical tests indicate an excess of rare frequency polymorphisms in the LL-Prochlorococcus data, but an excess of intermediate frequency polymorphisms in the HL-Prochlorococcus data. Distributions of the positively selected sites identified using the likelihood ratio test, when mapped onto the phycoerythrin tertiary structure, reveal that HL- and LL-phycoerythrin should be under different selective patterns. These findings may provide insights into the likely role of selection at the phycoerythrin locus and motivate further research to unveil the function of phycoerythrin in Prochlorococcus.

雨生红球藻中虾青素合成相关酶基因的转录调控序列分析

虾青素是一种具有极强抗氧化活性的类胡萝卜素，具有广泛的应用价值。雨生红球藻是一种单细胞绿藻，在逆境胁迫条件下能够大量合成并迅速积累虾青素，其积累量最高可达细胞干重的4%，从而成为目前首选的天然虾青素合成工具。但是，虾青素的大量积累总是发生在不适于生物量积累的环境胁迫条件下，虾青素积累与生物量积累之间成为一对矛盾，制约着虾青素大量生产。 异戊烯焦磷酸异构酶（IPI）、β-胡萝卜素酮化酶（BKT）和β-胡萝卜素氧化酶（CRTO）是虾青素合成过程中的相关酶。已有研究结果表明，这些酶基因的表达调控至少是部分发生在转录水平上的，这就为我们从转录水平上研究虾青素生物合成关键酶基因的调控机制提供了重要线索。 本文研究结果如下： 1. 利用基因组步移的方法克隆了两条异戊烯焦磷酸异构酶基因（ipi）的5’上游侧翼序列（1.8kb和2.5kb），预示着ipi的转录由不止一个启动子调控。 2. 利用上述方法克隆了两条β-胡萝卜素氧化酶基因（crtO）5'上游侧翼序列（1kb和2kb）。以lacZ为报告基因的瞬间表达实验结果表明，长度为320bp(-682/-363)的crtO 5'上游侧翼序列具有很强的启动转录活性，提示这段序列包含了启动子的结构。 3. 利用凝胶阻滞的方法研究了雨生红球藻中bkt强启动转录活性区域，即 309bp(-617/-309)启动子区域的转录因子结合位点，并发现在-396/-338的59bp区域内存在特异的核蛋白结合位点。通过序列分析，发现此59bp区域并不包含TATA或者CAAT-box，而是存在对光、缺氧、p-香豆酸及激素反应的G-box。 4. 根据国外已经获得的ipi和crtO全长cDNA序列，利用长距离PCR法从红球藻基因组中扩增到基因组序列。发现ipi和crtO均包含6个外显子和5个内含子，内含子的剪切位点基本符合GU-AG规律。并通过似然比检验（Likelihood ratio test）的方法发现两基因在进化过程中存在着正选择现象。 这些工作为下一步继续寻找与上述特定DNA调控区域特异结合的反式作用因子（蛋白质因子）奠定了基础。

The evaluation of evidence for exponentially distributed data

C.G.G. Aitken, Q. Shen, R. Jensen and B. Hayes. The evaluation of evidence for exponentially distributed data. Computational Statistics & Data Analysis, vol. 51, no. 12, pp. 5682-5693, 2007.

How Well Can TCP Infer Network State?

The Transmission Control Protocol (TCP) has been the protocol of choice for many Internet applications requiring reliable connections. The design of TCP has been challenged by the extension of connections over wireless links. We ask a fundamental question: What is the basic predictive power of TCP of network state, including wireless error conditions? The goal is to improve or readily exploit this predictive power to enable TCP (or variants) to perform well in generalized network settings. To that end, we use Maximum Likelihood Ratio tests to evaluate TCP as a detector/estimator. We quantify how well network state can be estimated, given network response such as distributions of packet delays or TCP throughput that are conditioned on the type of packet loss. Using our model-based approach and extensive simulations, we demonstrate that congestion-induced losses and losses due to wireless transmission errors produce sufficiently different statistics upon which an efficient detector can be built; distributions of network loads can provide effective means for estimating packet loss type; and packet delay is a better signal of network state than short-term throughput. We demonstrate how estimation accuracy is influenced by different proportions of congestion versus wireless losses and penalties on incorrect estimation.

A Bayesian Approach for TCP to Distinguish Congestion from Wireless Losses

(This Technical Report revises TR-BUCS-2003-011) The Transmission Control Protocol (TCP) has been the protocol of choice for many Internet applications requiring reliable connections. The design of TCP has been challenged by the extension of connections over wireless links. In this paper, we investigate a Bayesian approach to infer at the source host the reason of a packet loss, whether congestion or wireless transmission error. Our approach is "mostly" end-to-end since it requires only one long-term average quantity (namely, long-term average packet loss probability over the wireless segment) that may be best obtained with help from the network (e.g. wireless access agent).Specifically, we use Maximum Likelihood Ratio tests to evaluate TCP as a classifier of the type of packet loss. We study the effectiveness of short-term classification of packet errors (congestion vs. wireless), given stationary prior error probabilities and distributions of packet delays conditioned on the type of packet loss (measured over a larger time scale). Using our Bayesian-based approach and extensive simulations, we demonstrate that congestion-induced losses and losses due to wireless transmission errors produce sufficiently different statistics upon which an efficient online error classifier can be built. We introduce a simple queueing model to underline the conditional delay distributions arising from different kinds of packet losses over a heterogeneous wired/wireless path. We show how Hidden Markov Models (HMMs) can be used by a TCP connection to infer efficiently conditional delay distributions. We demonstrate how estimation accuracy is influenced by different proportions of congestion versus wireless losses and penalties on incorrect classification.

On classical and other methods of discriminant analysis and estimation of log-odds

For two multinormal populations with equal covariance matrices the likelihood ratio discriminant function, an alternative allocation rule to the sample linear discriminant function when n1 ≠ n2 ,is studied analytically. With the assumption of a known covariance matrix its distribution is derived and the expectation of its actual and apparent error rates evaluated and compared with those of the sample linear discriminant function. This comparison indicates that the likelihood ratio allocation rule is robust to unequal sample sizes. The quadratic discriminant function is studied, its distribution reviewed and evaluation of its probabilities of misclassification discussed. For known covariance matrices the distribution of the sample quadratic discriminant function is derived. When the known covariance matrices are proportional exact expressions for the expectation of its actual and apparent error rates are obtained and evaluated. The effectiveness of the sample linear discriminant function for this case is also considered. Estimation of true log-odds for two multinormal populations with equal or unequal covariance matrices is studied. The estimative, Bayesian predictive and a kernel method are compared by evaluating their biases and mean square errors. Some algebraic expressions for these quantities are derived. With equal covariance matrices the predictive method is preferable. Where it derives this superiority is investigated by considering its performance for various levels of fixed true log-odds. It is also shown that the predictive method is sensitive to n1 ≠ n2. For unequal but proportional covariance matrices the unbiased estimative method is preferred. Product Normal kernel density estimates are used to give a kernel estimator of true log-odds. The effect of correlation in the variables with product kernels is considered. With equal covariance matrices the kernel and parametric estimators are compared by simulation. For moderately correlated variables and large dimension sizes the product kernel method is a good estimator of true log-odds.