UMD Theses and Dissertations

Permanent URI for this collectionhttp://hdl.handle.net/1903/3

New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a given thesis/dissertation in DRUM.

More information is available at Theses and Dissertations at University of Maryland Libraries.

Browse

Subject: search.filters.subject.neural network

Search Results

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Causal Survival Analysis – Machine Learning Assisted Models: Structural Nested Accelerated Failure Time Model and Threshold Regression
    (2022) Chen, Yiming; Lee, Mei-Ling ML; Mathematical Statistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Time-varying confounding for intervention complicates causal survival analysis when the data are collected in a longitudinal manner. Traditional survival models that only adjust for time-dependent covariates provide a biased causal conclusion for the intervention effect. Some techniques have been developed to address this challenge. Nevertheless, these existing methods may still lack power, and suffer from computational burden given high dimensional data with a temporally connected nature. The first part of this dissertation focuses on one of the methods that deal with time-varying confounding, the Structural Nested Model and associated G-estimation. Two Neural Networks (GE-SCORE and GE-MIMIC) were proposed to estimate the Structural Nested Accelerated Failure Time Model. The proposed algorithms can provide less biased and individualized intervention causal effect estimation. The second part explored the causal interpretations and applications of the First-Hitting-Time based Threshold Regression Model using a Wiener process. Moreover, a Neural Network expansion of this specific type of Threshold Regression (TRNN) was explored for the first time.
  • Thumbnail Image
    Item
    DATA-DRIVEN STUDIES OF TRANSIENT EVENTS AND APERIODIC MOTIONS
    (2019) Wang, Rui; Balachandran, Balakumar; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    The era of big data, high-performance computing, and machine learning has witnessed a paradigm shift from physics-based modeling to data-driven modeling across many scientific fields. In this dissertation work, transient events and aperiodic motions of complex nonlinear dynamical system are studied with the aid of a data- driven modeling approach. The goal of the work has been to further the ability for future behavior prediction, state estimation, and control of related behaviors. It is shown that data on extreme waves can be used to carry out stability analysis and ascertain the nature of the transient phenomenon. In addition, it is demonstrated that a low number of soliton elements can be used to realize a rogue wave on the basis of nonlinear interactions amongst the basic elements. The pro- posed nonlinear phase interference model provides an appealing explanation for the formation of ocean extreme wave and related statistics, and a superior reconstruction of the Draupner wave event than that obtained on the basis of linear superposition. Chaotic data, another manifestation of aperiodic motions, which are obtained from prototypical ordinary differential and partial differential systems are considered and a neural machine is realized to predict the corresponding responses based on a limited training set as well to forecast the system behavior. A specific neural architecture, called the inhibitor mechanism, has been designed to enable chaotic time series forecasting. Without this mechanism, even the short-term predictions would be intractable. Both autonomous and non-autonomous dynamical systems have been studied to demonstrate the long-term forecasting possibilities with the de- veloped neural machine. For each dynamical system considered in this dissertation, a long forecasting horizon is achieved with a short historical data set. Furthermore, with the developed neural machine, one can relax the requirement of continuous historical data measurements, thus, providing for a more pragmatic approach than the previous approaches available in the literature. It is expected that the efforts of this dissertation work will lead to a better understanding of the underlying mechanism of transient and aperiodic events in complex systems and useful techniques for forecasting their future occurrences.
  • Thumbnail Image
    Item
    Extracting Symbolic Representations Learned by Neural Networks
    (2012) Huynh, Thuan Quang; Reggia, James A; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Understanding what neural networks learn from training data is of great interest in data mining, data analysis, and critical applications, and in evaluating neural network models. Unfortunately, the product of neural network training is typically opaque matrices of floating point numbers that are not obviously understandable. This difficulty has inspired substantial past research on how to extract symbolic, human-readable representations from a trained neural network, but the results obtained so far are very limited (e.g., large rule sets produced). This problem occurs in part due to the distributed hidden layer representation created during learning. Most past symbolic knowledge extraction algorithms have focused on progressively more sophisticated ways to cluster this distributed representation. In contrast, in this dissertation, I take a different approach. I develop ways to alter the error backpropagation neural network training process itself so that it creates a representation of what has been learned in the hidden layer activation space that is more amenable to existing symbolic representation extraction methods. In this context, this dissertation research makes four main contributions. First, modifications to the backpropagation learning procedure are derived mathematically, and it is shown that these modifications can be accomplished as local computations. Second, the effectiveness of the modified learning procedure for feedforward networks is established by showing that, on a set of benchmark tasks, it produces rule sets that are substantially simpler than those produced by standard backpropagation learning. Third, this approach is extended to simple recurrent networks, and experimental evaluation shows remarkable reduction in the sizes of the finite state machines extracted from the recurrent networks trained using this approach. Finally, this method is further modified to work on echo state networks, and computational experiments again show significant improvement in finite state machine extraction from these networks. These results clearly establish that principled modification of error backpropagation so that it constructs a better separated hidden layer representation is an effective way to improve contemporary symbolic extraction methods.
  • Thumbnail Image
    Item
    Using machine learning to measure the cross section of top quark pairs in the muon+jets channel at the Compact Muon Solenoid
    (2011) Kirn, Malina Aurelia; Hadley, Nicholas; Applied Mathematics and Scientific Computation; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    The cross section for pp to top-antitop production at a center of mass energy of 7 TeV is measured using a data sample with integrated luminosity 36.1 inverse pb collected by the CMS detector at the LHC. The analysis is performed on a computing grid. Events with an isolated muon and three hadronic jets are analyzed using a multivariate machine learning algorithm. Kinematic variables and b tags are provided as input to the algorithm; output from the algorithm is used in a maximum likelihood fit to determine top-antitop event yield. The measured cross section is 151 +/- 15(stat.) +35/-28(syst.) +/- 6(lumi.) pb.
  • Thumbnail Image
    Item
    IN-SITU MEASUREMENT OF EPITHELIAL TISSUE OPTICAL PROPERTIES: DEVELOPMENT AND IMPLEMENTATION OF DIFFUSE REFLECTANCE SPECTROSCOPY TECHNIQUES
    (2009) Wang, Quanzeng; Wang, Nam Sun; Chemical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Cancer is a severe threat to human health. Early detection is considered the best way to increase the chance for survival. While the traditional cancer detection method, biopsy, is invasive, noninvasive optical diagnostic techniques are revolutionizing the way that cancer is diagnosed. Reflectance spectroscopy is one of these optical spectroscopy techniques showing promise as a diagnostic tool for pre-cancer detection. When a neoplasia occurs in tissue, morphologic and biochemical changes happen in the tissue, which in turn results in the change of optical properties and reflectance spectroscopy. Therefore, a pre-cancer can be detected by extracting optical properties from reflectance spectroscopy. This dissertation described the construction of a fiberoptic based reflectance system and the development of a series of modeling studies. This research is aimed at establishing an improved understanding of the optical properties of mucosal tissues by analyzing reflectance signals at different wavelengths. The ultimate goal is to reveal the potential of reflectance-based optical diagnosis of pre-cancer. The research is detailed in Chapter 3 through Chapter 5. Although related with each other, each chapter was designed to become a journal paper ultimately. In Chapter 3, a multi-wavelength, fiberoptic system was constructed, evaluated and implemented to determine internal tissue optical properties at ultraviolet A and visible wavelengths. A condensed Monte Carlo model was deployed to simulate light-tissue interaction and generate spatially distributed reflectance data. These data were used to train an inverse neural network model to extract tissue optical properties from reflectance. Optical properties of porcine mucosal and liver tissues were finally measured. In Chapter 4, the condensed Monte Carlo method was extended so that it can rapidly simulate reflectance from a single illumination-detection fiber thus enabling the calculation of large data sets. The model was implemented to study spectral reflectance changes due to breast cancer. The effect of adding an illumination-detection fiber to a linear array fiber for optical property determination was also evaluated. In Chapter 5, an investigation of extracting the optical properties from two-layer tissues was performed. The relationship between spatially-resolved reflectance distributions and optical properties in two-layer tissue was investigated. Based on all the aforementioned studies, spatially resolved reflectance system coupled with condensed Monte Carlo and neural network models was found to be objective and appear to be sensitive and accurate in quantitatively assessing optical property change of mucosal tissues.