A human listener's ability to organize the time-frequency (T-F) energy of the same sound source into a single stream is termed auditory scene analysis (ASA). Computational auditory scene analysis (CASA) seeks to organize sound based on ASA principles. This dissertation presents a systematic effort on sequential organization in CASA. The organization goal is to group T-F segments from the same speaker that are separated in time into a single stream. This dissertation proposes a speaker-model-based sequential organization framework and it shows better grouping performance than feature-based methods. Specifically, a computational objective is derived for sequential grouping in the context of speaker recognition for multi-talker mixtures. This formulation leads to a grouping system that searches for the optimal grouping of separated speech segments. A hypothesis pruning method is then proposed that significantly reduces search space and time while achieving performance close to that of exhaustive search. Evaluations show that the proposed system improves both grouping performance and speech recognition accuracy. The proposed system is then extended to handle multi-talker as well as non-speech intrusions using generic models. The system is further extended to deal with noisy inputs from unknown speakers. It employs a speaker quantization method that extracts generic models from a large speaker space. The resulting grouping performance is only moderately lower than that with known speaker models. In addition, this dissertation presents a systematic effort in robust speaker recognition. A novel usable speech extraction method is proposed that significantly improves recognition performance. A general solution is proposed for speaker recognition under additive-noise conditions. Novel speaker features are derived from auditory filtering, and are used in conjunction with an uncertainty decoder that accounts for mismatch introduced in CASA front-end processing. Evaluations show (open full item for complete abstract)

In our research, we exploit the relationship between properties of U-statistics and Energy statistics (V-statistics) to come up with non-parametric tests in change-point analysis. \cite{lee} provided a wide discussion on asymptotic behaviour, and connection between U-statistics and V-statistics when large samples. Many other researchers such as \cite{sen1974}, \cite{serfling1980} and \cite{neuhaus1977} studied connections between U-statistic and V-statistics and their asymptotic properties. We first propose a non-parametric test to detect change in the distribution based on MIC using energy statistics. The proposed energy-statistic based MIC is used for model selection between null and alternative hypothesis models. We achieve this by adopting the idea of the works of \cite{chen} and \cite{pan} and apply energy distance statistic. To test the performance of our proposed test, we assess the finite sample properties and compare efficiency and powers of different methods with that of our method. We then discuss applications of our proposed test in two different real-life examples for detecting change in mean and variance, respectively. Since the underlying distribution is unknown, we use bootstrap approximations for the p-values as proposed by \cite{hangfen2009} in detecting unknown change points in means and variances. In the second part of my dissertation, we propose a non-parametric sequential test based on energy statistics \cite{rizzo2013} to detect changes in distribution for independent random variables. In their study, \cite{Oscar} considered backward-looking windows each of length $L$ across the pooled data, and then retrospectively investigate if there is evidence for a change point between the times $\text{max}\{t-L,1\}$ and $t$, for any given time $t$. We adopt this idea to come up with a test statistic similar in structure based on energy statistics. We compare the performance of this method in terms of false-alarm rates and powers to existing sequenti (open full item for complete abstract)

Sequential change point analysis aims to detect structural change as quickly as possible when the process state changes. A good sequential change point detection procedure is expected to minimize the detection delay time and the risk of raising false alarm. Existing sequential change point detection methods cannot be applicable for high-dimensional data because they are univariate in nature and thus present challenges. In the first part of the dissertation, we develop a monitoring method to detect structural change in smoothly clipped absolute deviation (SCAD) penalized regression model for high-dimensional data after the historical sample with the sample size m. The unknown pre-change regression coefficients are replaced by the SCAD penalized estimator. The asymptotic properties of the proposed test statistics are derived. We conduct a simulation study to evaluate the performance of the propose method. The proposed method is applied to the gene expression in the mammalian eye data to detect changes sequentially. In the second part of the dissertation, we develop a sequential change point detection method to monitor structural changes in SACD penalized quantile regression (SPQR) model for high-dimensional data. We derive the asymptotic distributions of the test statistic under the null and alternative hypotheses. Furthermore, to improve the performance of the SPQR method, we propose the Post-SCAD penalized quantile regression estimator (P-SPQR) for high-dimensional data. Simulations are conducted under different scenarios to study the finite sample properties of the SPQR and P-SPQR methods. A real data application is provided to demonstrate the effectiveness of the method. In the third and fourth part of the dissertation, we investigate the change point problem for Skew-Normal distribution and three parameter Weibull distribution respectively. Besides detecting and obtaining the point estimate of a change location, we propose an estimation procedure based (open full item for complete abstract)

Sequential sampling models such as the Diusion Decision Model (DDM) and the Linear Ballistic Accumulator (LBA) are often used as measurement tools in psychology. However, two practical issues regarding the use of them have received limited attention: Identiabilities of the models and the appropriateness of the follow-up testing procedures in terms of statistical power. In the present research, I address these problems to ll the gap in the literature. Specically, I do the following two things. First, I formally conduct identiability analyses of DDM and LBA. As a result, I argue that some version of DDM, namely the "full DDM," is unidentiable, even when multiple experimental conditions are employed. I show that this problem arises due to the excess flexibility of the model, and it can only be solved by reducing the number of free parameters to be estimated. Second, I demonstrate that the use of t-tests while comparing parameter estimates cannot be justied because such a practice assumes an over-simplied, single-level hierarchical model. As such, the statistical power is shown to be suboptimal. Instead, it is recommended that one employ an alternative procedure that explicitly models uncertainties about the parameter estimates, such as meta-regression or Hierarchical Bayes (HB). It is shown that such solutions are better theoretically grounded, exhibit larger statistical power, or yield more precise parameter estimates. Recommendations for substantive researchers are provided based on these considerations.

For many years, reverse engineering of hardware designs has been an area of great interest. Efficient and structured analysis of fabricated designs is important for several reasons, such as design validation, IP protection, process quality control etc. More recently, multivariate nature of semiconductor supply-chain has opened doors for insertion of obscure hardware vulnerabilities making hardware integrity check essential for ICs used in critical application areas. Most traditional hardware reverse engineering techniques are invasive and lead to a partial or complete destruction of the system under investigation which is often times unwanted. In this thesis, we present scalable, non-invasive procedures to reverse engineer unknown CMOS based ICs. Specifically, the focus is on black-box analysis of unknown Moore Finite State Machine based sequential circuit designs. We present two different recovery techniques based on a novel analysis approach that combines investigation of input-output responses and power consumption of the system under investigation. The first technique performs a tree-based guided exploration of the machine structure and employs subtree matching to identify distinct and equivalent states. The second technique translates machine exploration and state identification into a constraint satisfaction problem that can be efficiently handled by a SMT Solver. The advantage of the tree-based approach is that it guarantees a minimally equivalent recovery, whereas the solver-based approach works adaptively and hence, faster and scalable to handle large machines. Both these techniques successfully recover a logically equivalent state machine structure. To study the efficiency and performance of the proposed techniques we present its implementation. We compare the execution times for different standard MCNC benchmark machines and show that the solver-based recovery technique is faster.

This study investigates the forms of social organization found in modern classrooms and classroom lessons, as they are encountered by children in a rural Turkish village, in its village school. It is a study of the early childhood education and educational experience of these children. It is composed of a collection of ethnographic descriptions and discourse analyses that examine the social–organizational forms of life found in the village, those found in the school, and the experiences and attitudes of village children, parents and elders towards schooling and its place in their lives. These three foci organize the early childhood educational experiences of these children. They organize a larger picture of their encounters with schooling in the early grades and the place of schooling in the lives of their families. All children develop the competencies required of them as they participate in their routine daily activities with little, if any, direct or explicit instruction. Much of early childhood education is of this implicit, participatory character. The instruction that is implicit to their participation, whether in the classroom community or the village community, is both taken granted, and closely studied by the children. It is in these ways that modern school rooms engage in the production of modern students, among village children.

UCII has been involved with Bridge Monitoring for over two decades and a Bridge Health Monitoring System for the Ironton – Russell Bridge was developed and is being maintained by UCII. The Bridge Monitoring System collects data from the sensors placed on the bridge and sends real time updates regarding any abnormal behavior. Whenever an abnormal behavior is detected, `Alarms" are sent out via email and in order to be better understand the alarm scenario, the feature "Active Connection" was developed. Active Connection involves establishing a dynamic connection to the system on the bridge which makes two way communication possible to get additional data points when the Alarms occur. By utilizing this connection, additional data points are then obtained which are analyzed using sequential analysis and a severity level for the alarm is declared. This feature of active connection is helpful in better assessing the alarm situation and improving the reliability of the bridge monitoring system on the Ironton – Russell Bridge.

The sequential linear model is widely employed to analyze the dynamic data where the response variable at each time point incorporates the lagged results from the previous time point. With the lagged dependent response variables added to the model longitudinally, the issue of multicollinearity arises. In such situations, the Lasso method proposed by Tibshirani (1996) addresses both parameter estimation and variable selection simultaneously. However, in high-dimensional data and multicollinearity, the Lasso method can introduce bias in coefficient estimation and inconsistency in variable selection. To improve the Lasso method, a number of different penalty terms are proposed. Among the Lasso methods with different penalty terms, selecting an appropriate estimation and variable selection method is challenging work because it requires balancing the trade-off between achieving low bias and maintaining high prediction accuracy. One of the primary inferences in the sequential linear model is to predict the response variable with high accuracy and relatively minimal prediction errors, thereby saving time and expenses. To achieve this goal, we propose the estimation and variable selection method based on the Lasso, named Smoothly Clipped Absolute Deviation Penalty (SCAD) (Fan and Li, 2001), in the sequential linear model. The proposed SCAD method performs effectively in parameter estimation with low bias and variable selection with low predicted errors. In the demonstration of the effectiveness of the proposed method, we conduct the simulations where we compare the SCAD method with other methods including the ordinary least squares (OLS), Lasso, and Adaptive Lasso in both linear regression and sequential linear models. Since time series refers to a sequence of data generated at each time point, where the lagged response variable at each time point is used as a predictor in the subsequent time point model, accounting for errors based on assumptions, we simulate the data in (open full item for complete abstract)

Sequential change-point analysis identifies a change of probability distribution in an infinite sequence of observations generated by a process, by repetitively performing a hypothesis test each time a new observation is generated and added to the current data set. It has important applications in many fields, such as financial investment, system monitoring, and quality control. While lots of research have been done for different scenarios, especially time series, few works have been developed for skew data, as well as for the case where the distribution family of observations is unspecified. Hence, in this dissertation, we focus on developing a sequential point detection procedure for the skew-normal distribution family, and a nonparametric procedure based on the modification of previous methods. In the first part of the dissertation, we propose a sequential change-point detection rule for skew-normal distribution, by modifying the procedures proposed by Mei (2006). We focus on the change of location and shape parameters, respectively under the simple and composite alternative hypothesis. We derive the optimality of our modified procedure for location parameter under a simple alternative hypothesis. Also, the simulation shows that our new procedure has fewer false alarms than previous methods when the exact value of pre-change and post-change parameters are not specified. In the second part, we proposed a nonparametric sequential change-point detection procedure, by modifying Page's CUSUM procedure and the well-known Shiryaev-Roberts (SR) procedure. More specifically, we substitute the parametric likelihood function in the two methods with empirical likelihood (EL), which allows us to perform a likelihood ratio test without knowing the distribution family. Also, we assume training data are available to estimate pre-change and post-change parameters. Different versions of empirical likelihood are applied and simulations are conducted to show their perfo (open full item for complete abstract)

We consider a sequence of i.i.d. random observations from a standard skew-normal distribution with skewness λ, denoted as SN(λ). We wish to test H0: λ = λ0 vs H1: λ = λ1, where λ0 and λ1 (λ0 = λ1) are unknown constants, with target type-I and type-II error probabilities, denoted as α and β respectively. We first describe some interesting characteristics of the skew-normal distribution and then adopt Wald's sequential probability ratio test (SPRT) to perform the decision making and determine, on average, how many observations are needed to make such a decision. We choose numerous values of λ0 and λ1 to study how the chosen values affect the average sample number (ASN). We then compare these theoretical average sample numbers to those obtained through simulations. The approximations developed are applied to a set of BMI data. We develop a nonparametric goodness-of-fit test for the hypothesis H0: F = SN(μ, ς, λ) vs H1: F = SN(μ, ς, λ), based on the energy distance where F is the distribution of X1, ... , Xn. We first describe the energy distance and functions of energy distance, called energy statistics, along with some useful properties. We also briefly describe currently available goodness-of-fit tests for the skew-normal distribution in order to make comparisons with the proposed test. Simulations are conducted to indicate that the proposed test controls the Type-I error rate well and power studies show a higher detection rate for skew-normal than existing tests. The proposed test is applied to a set of IQ data and to the BMI data from Chapter 2. We develop a nonparametric goodness-of-fit test for the hypothesis H0: F = SEP(μ,ς, λ,ν) vs H1: F = SEP(μ, ς, λ,ν) where SEP(μ, ς, λ,ν) is the skewed exponential power distribution. This proposed test is based on the energy distance described in Chapter 3. We first describe the exponential power distribution in its symmetric version first while discussing some useful properties. Then we place the skewness parameter (open full item for complete abstract)

Sensors capable of collecting wide area motion imagery (WAMI), video synthetic aperture radar (SAR), and other high frame rate sensor modalities provide massive amounts of high-resolution data. Such data allows for the use of multiple images in exploitation tasks which may have traditionally used single images or single pairs of images. One such task is change detection. This dissertation presents new statistical methods for change detection that provide for the exploitation of multiple images per pass. Uncertainty in image registration can degrade change detection performance. Registration accuracy is analyzed, and the impact of registration uncertainty is propagated to the registered imagery. A statistical understanding of this uncertainty is incorporated into the sequential change detection algorithm to mitigate performance degradation due to registration errors. Theoretical results are verified through simulation experiments and with measured data sets.

Multilocus calculations using all available information on all pedigree members are important for linkage analysis. Exact calculation methods in linkage analysis are limited in either the number of loci or the number of pedigree members they can handle. In this thesis, we propose a Monte Carlo method for linkage analysis based on sequential imputation. Unlike exact methods, sequential imputation can handle both a moderate number of loci and a large number of pedigree members. Sequential imputation does not have the problem of slow mixing encountered by Markov chain Monte Carlo methods because of high correlation between samples from pedigree data. This Monte Carlo method is an application of importance sampling in which we sequentially impute ordered genotypes locus by locus and then impute inheritance vectors conditioned on these genotypes. The resulting inheritance vectors together with the importance sampling weights are used to derive a consistent estimator of any linkage statistic of interest. The linkage statistic can be parametric or nonparametric; we focus on nonparametric linkage statistics. We showed that sequential imputation can produce accurate estimates within reasonable computing time. Then we performed a simulation study to illustrate the potential gain in power using our method for multilocus linkage analysis with large pedigrees. We also showed how sequential imputation can be used in haplotype reconstruction, an important step in genetic mapping. In all of the applications of sequential imputation we can incorporate interference, which often is ignored in linkage analysis due to computational problems. We demonstrated the effect of interference on haplotyping and linkage analysis. We have implemented sequential imputation for multilocus linkage analysis in a user-friendly software package called SIMPLE (Sequential Imputation for Multi-Point Linkage Estimation). SIMPLE currently can estimate LOD scores, IBD sharing statistics and haplotype configur (open full item for complete abstract)