Defense Notices

With an increasing trend of adoption of machine learning in various real-world problems, the need for transparent and reliable models has become apparent. Especially in some socially consequential applications, such as medical diagnosis, credit scoring, and decision making in educational systems, it may be problematic if humans cannot understand and trust those models. To this end, in this work, we propose a novel machine learning algorithm, constructivism learning. To achieve transparency, we formalized a Bayesian nonparametric approach using sequential Dirichlet Process Mixture of prediction models to support constructivism learning. To achieve reliability, we exploit two strategies, reducing model uncertainty and increasing task construction stability by leveraging techniques in active learning and self-paced learning. 

Distance metrics are concerned with learning how objects are similar, and are a critical component of many machine learning algorithms such as k-nearest neighbors and kernel machines. Traditional metrics are unable to adapt to data with heterogenous interactions in the feature space. State of the art methods consider learning multiple metrics, each in some way local to a portion of the data. Selecting how the distance metrics are local to the data is done apriori, with no known best approach. 
In this proposal, we address the local metric learning scenario from three complementary perspectives. In the first direction, we consider a spatial approach, and develop an efficient Frank-Wolfe based technique to learn local distance metrics directly in a high-dimensional input space. We then consider a view-local perspective, where we associate each metric with a separate view of the data, and show how the approach naturally evolves into a multiple kernel learning problem. Finally, we propose a new function for learning a metric which is based on a newly discovered operator called the t-product, here we show that our metric is composed of multiple parts, with each portion local to different interactions in the input space. 

Resource limited embedded systems provide a great challenge to programming using functional languages. Although we cannot program these embedded systems directly with Haskell, we show than an embedded domain specific language is able to be used to program them, providing a user friendly environment for both prototyping and full development. The Arduino line of microcontroller boards provide a versatile, low cost and popular platform for development of these resource limited systems, and we use this as the platform for our DSL research. 

First we provide a shallowly embedded domain specific language and a firmware interpreter, allowing the user to program the Arduino while tethered to a host computer. Second, we add a deeply embedded version, allowing the interpreter to run standalone from the host computer, as well as allowing us to compile the code to C and then machine code for efficient operation. Finally, we develop a method of transforming the shallowly embedded DSL syntax into the deeply embedded DSL syntax automatically.

With the promise of meeting future capacity demands for mobile broadband communications, 3D massive-MIMO/Full Dimension MIMO (FD-MIMO) systems have gained much interest among the researchers in recent years. Apart from the huge spectral efficiency gain offered by the system, the reason for this great interest can also be attributed to significant reduction of latency, simplified multiple access layer, and robustness to interference. However, in order to completely extract the benefits of massive-MIMO systems, accurate channel state information is critical. In this thesis, a channel estimation method based on direction of arrival (DoA) estimation is presented for massive- MIMO OFDM systems. To be specific, the DoA is estimated using Estimation of Signal Parameter via Rotational Invariance Technique (ESPRIT) method, and the root mean square error (RMSE) of the DoA estimation is analytically characterized for the corresponding MIMO-OFDM system.

Billions of dollars are lost every year to successful cyber attacks that are fundamentally enabled by software vulnerabilities. Modern cyber attacks increasingly threaten individuals, organizations, and governments, causing service disruption, inconvenience, and costly incident response. Given that such attacks are primarily enabled by software vulnerabilities, this work examines the efficacy of using change metrics, along with architectural burst and maintainability metrics, to predict modules and files that should be analyzed or tested further to excise vulnerabilities prior to release. 

The problem addressed by this research is the residual vulnerability problem, or vulnerabilities that evade detection and persist in released software. Many modern software projects are over a million lines of code, and composed of reused components of varying maturity. The sheer size of modern software, along with the reuse of existing open source modules, complicates the questions of where to look, and in what order to look, for residual vulnerabilities. 

Traditional code complexity metrics, along with newer frequency based churn metrics (mined from software repository change history), are selected specifically for their relevance to the residual vulnerability problem. We compare the performance of these complexity and churn metrics to architectural level change burst metrics, automatically mined from the git repositories of the Mozilla Firefox Web Browser, Apache HTTP Web Server, and the MySQL Database Server, for the purpose of predicting attack prone files and modules. 

We offer new empirical data quantifying the relationship between our selected metrics and the severity of vulnerable files and modules, assessed using severity data compiled from the NIST National Vulnerability Database, and cross-referenced to our study subjects using unique identifers defined by the Common Vulnerabilities and Exposures (CVE) vulnerability catalog. Our results show that architectural level change burst metrics can perform well in situations where more traditional complexity metrics fail as reliable estimators of vulnerability severity. In particular, results from our experiments on Apache HTTP Web Server indicate that architectural level change burst metrics show high correlation with the severity of known vulnerable modules, and do so with information directly available from the version control repository change-set (i.e., commit) history. 

Incomplete and numerical data are common for many application domains. There have been many approaches to handle missing data in statistical analysis and data mining. To deal with numerical data, discretization is crucial for many machine learning algorithms. However, most of the discretization algorithms cannot be applied to incomplete data sets. 

Multiple Scanning is an entropy based discretization method. Previous research shown it outperforms commonly used discretization methods: Equal Width or Equal Frequency discretization. In this work, Multiple Scanning is tested on C4.5 and MLEM2 on incomplete datasets. Results show for some data sets, the setup utilizing Multiple Scanning as preprocessing performs better, for the other data sets, C4.5 or MLEM2 should be used by themselves. Our conclusion is that there are no universal optimal solutions for all data sets. Setup should be custom-made. 

Rough set theory is a useful approach for decision rule induction, which is applied, to large life data sets. Lower and upper approximations of concepts values are used to induce rules for incomplete data sets. In our research we will study validity of modifications suggested to characteristic relation. We discuss the implementation of modifications to characteristic relation, and the local definability of each modified set. We show that all suggested modifications sets are not locally definable except for maximal consistent blocks that are restricted to data set with “do not care” conditions. A comparative analysis was conducted for characteristic sets and modifications in terms of cardinality of lower and upper approximations of each concept and decision rules induced by each modification. In this thesis, experiments were conducted on four incomplete data sets with lost and “do not care “ conditions. LEM2 algorithm was implemented to induce certain and possible rules form the incomplete data set. To measure the classification average error rate for induced rules, ten-fold cross validation was implemented. Our results show that there is no significant difference between the qualities of rule induced from each modification.

Increasing the rate of crop yield is an important issue to meet projected future crop production demands. Breeding efforts are being made to rapidly improve crop yields and make them more stress-resistance. Accelerated molecular breeding techniques, in which desirable plant physical traits are selected based on genetic markers, rely on accurate and rapid methods to link plant genotypes and phenotypes. Advances in next-generation-DNA sequencing have made genotyping a fast and efficient process. In contrast, methods for characterizing physical traits remain inefficient. 
The height of wheat crop is an important trait as it may be related to yield and biomass. It is also an indicator of plant growth-stage. Recent high-throughput-phenotyping experiments have used sensing techniques to measure canopy height based on ultrasound sonar and cameras. The main drawback of these methods is that the ground topography is not directly measured. 
In contrast to current sensors, ultra-wideband radars have the potential to take distance measurements to the top of the canopy and the ground simultaneously. We propose the study of ultra-wideband radar for measuring wheat crop heights. Specifically, we propose to study the effects of canopy constituents on the ranging radar-return or impulse-response, as well as on the frequency-response. First, a numerical simulator will be developed to accurately calculate the radar response at different canopy conditions. Second, a parametric study will be performed with aforementioned simulator. Lastly, an estimation algorithm for crop canopy heights will be developed, based on the parametric study. 

Energy harvesting from solar sources in an attempt to increase efficiency has sparked interest in many communities to develop more energy harvesting applications for renewable energy topics. Advanced technical methods are required to ensure the maximum available power is harnessed from the photovoltaic (PV) system. This work proposes a new discrete-in-time extremum-seeking based technique for tracking the maximum power point of a photovoltaic array. The proposed method is a true maximum power point tracker that can be implemented with reasonable processing effort on an expensive digital controller. The approach is to study the stability analysis of the proposed method to guarantee the convergence of the algorithm. The proposed method should exhibit better performance in comparison to conventional Maximum Power Point Tracking (MPPT) methods and require less computational effort than the complex mathematical methods. 

Online social networks (OSNs) like Twitter provide an open platform for users to easily convey their thoughts and ideas from personal experiences to breaking news. With the increasing popularity of Twitter and the explosion of tweets, we have observed large amounts of potentially sensitive/private messages being published to OSNs inadvertently or voluntarily. The owners of these messages may become vulnerable to online stalkers or adversaries, and they often regret posting such messages. Therefore, identifying tweets that reveal private/sensitive information is critical for both the users and the service providers. However, the definition of sensitive information is subjective and different from person to person. To develop a privacy protection mechanism that is customizable to fit the needs of diverse audiences, it is essential to accurately and automatically identify and classify potentially sensitive tweets. 
In this project, we adopted a two-step approach - private tweet identification, and private tweet classification. We make the first attempt to classify private tweets into two main categories, sensitive and nonsensitive - private tweet identification, followed by private tweet classification where we categorize the sensitive tweets into 13 pre-defined topics. We consider private tweet identification and private tweet classification as dual-problems. Progress towards one of them will eventually benefit the other. We used a 2-layer classification approach, where we explore different combinations of classifiers, and analyze the performance of each combination.