This paper addresses the high model complexity and overconfident frame labeling of state-of-the-art (SOTA) action segmenters. Their complexity is typically justified by the need to sequentially refine action segmentation through multiple stages of a deep architecture. However, this multistage refinement does not take into account uncertainty of frame labeling predicted...
The use of genetic algorithms to compose music and generate sounds is an area of interest in the artificial intelligence field. Music and instrument sounds have known rules and structures that can be followed which make them well-suited for genetic algorithms. However, genetic algorithms still struggle to generate sounds comparable...
Over 37,000 people die each year in automobile accidents, with many of these fatalities resulting from collisions with emergency vehicles. The rise of autonomous cars creates the need for an accurate and failsafe method of detecting and responding to emergency vehicles safely and on time. This thesis investigates the ability...
Autonomous robotic agents are on their way to becoming in-home personal assistants, construction assistants, and warehouse workers. The degree of autonomy of such systems is reflected by the manner in which we specify goals to them; the abstraction of low-level commands to high-level goals goes hand-in-hand with increased autonomy. In...
Learning to recognize objects is a fundamental and essential step in human perception and understanding of the world. Accordingly, research of object discovery across diverse modalities plays a pivotal role in the context of computer vision. This field not only contributes significantly to enhancing our understanding of visual information but...
While digital inclusivity researchers and software practitioners have been trying to address exclusion biases in Windows, Icons, Menus, and Pointers (WIMP) user interfaces (UIs) for a long time, little has been done to investigate if and how inclusive software design and its methods that have been devised for WIMP UIs...
A video server is the multimedia equivalent of a data file server. As video goes digital video servers that support digital media are required for production and broadcasting in television companies. Grass Valley Group provides comprehensive digital video solutions through a range of video servers called Profiles.
The Profile Media...
Structure Query Language (SQL) is widely used to access data stored in relational database systems. Although a powerful and flexible language, SQL can also be complex and hard to learn. For most new SQL users, it's easy to write SQL statement by following SQL grammar and syntax rules, but it's...
We have developed a framework for Web-based GIS/database applications which allow users to insert, update, delete, and query data with a map interface displayed by Web browsers. The framework was designed so that a Web-based GIS application that uses ArcIMS as a map server can be easily created, customized, and...
Web-Site Generator 3 (WebSiteGen3) is a rapid application development (RAD) tool that generates ASP.NET forms to insert, query, update and delete data stored in the user tables in a SQL Server 2000 database. WebSiteGen3 uses a graphical user interface to show the user tables in a hierarchical tree based on...
Analysis of programs forms an important activity in the field of software engineering. It is necessary to help understand the code, which facilitates comprehensive testing, maintenance and optimization of code. Aristotle is a tool for analyzing programs written in C. We have designed a system on similar lines for Java...
I present a new heuristic search approach to compute approximate answers for the probability query in belief nets. This approach can compute the 'best' bounds for a query in a period of any given time (if time permitted, it will get an exact value). It inherits the essence of Symbolic...
Wireless Application Protocol (WAP) is an industry standard aimed to bring the web to handheld devices. The handheld devices are constrained by battery life and it becomes important that power is conserved across web transactions. The power consumed by the handheld device is directly proportional to the time taken for...
We have developed a prototype web-based GIS application for tracking the locations of moving entities. This application, which is called the Responder application, consists of two parts: the Responder client and the Responder server. The Responder client is a .NET application written in C#. It reads the location data from...
Global Positioning Systems have allowed for precise timing of power system measurements over wide areas. This newly found capability has the potential to provide much greater insight into the operation of the power system and its response to contingencies, but few analytical techniques currently exist that provide enough robustness and...
With sequential computing technology reaching its speed limits, parallel processing is emerging as the key to very-high-speed computation. However, developing a parallel program is by no means a simple task; neither is analyzing the performance of parallel programs.
C* is a high-level data-parallel language that hides explicit message passing and...
High level data-parallel languages are easy to use and shield the programmer from machine specific details. A simple and efficient way of providing an interface to such languages is to develop a machine-independent compiler and a routing library, which isolates the low-level machine dependent communication functions, The compiler translates the...
Component based software technologies are viewed as essential for creating the software systems of the future. However, the use of externally provided component has serious drawbacks for a wide range of software engineering activities often because of a lack of information about the components. One such drawback involves validation of...
Web-based Pesticide Screening Tool (Web-PST) is a web-based software application for evaluating the potential risk of pesticides on the surrounding environment. It uses the formulas and standards specified by the Soil/Pesticide Interaction Screening Procedure Version II (SPISP II). Web-PST closely models the stand-alone Windows application Windows Pesticide Screening Tool (WIN-PST)....
Dataparallel C has been designed for various kinds of architectures and was developed jointly at University of New Hampshire and Oregon State University.
This project dealt with porting the libraries to the nCUBE 2 system and helping a user compile his programs on nCUBE 2 directly from a Sun workstation...
CREEDA-BG (Crop Rotation Economic and Environmental Decision-Aid Budget Generator) is a web-based enterprise budgeting tool that allows a user to manage information on her fields, rotations, crops, and operations, and to view estimates of income and expenses. The user can then use this information to evaluate alternative plans and make...
In this paper we describe a fuzzy logic based control system implemented on a PC architecture. The rule based inference engine of this expert system is easily configured through an ascii text file and is demonstrated to be capable of controlling various simulations, including an inverted pendulum. The controller is...
Component based software technologies are viewed as essential for creating the software systems of the future. However the use of externally provided components has serious drawbacks for a wide range of software engineering activities often because of a lack of information about the components. One such drawback involves validation of...
We present a model for a distributed virtual market place that can be constructed on the Internet to support selling and buying requests, such as those found as classified advertisements. One requirement for a transaction to take place in the virtual market place is that a sell request and a...
We have developed SearchPak, a machine independent parallel searching tool on shared and distributed memory machines. It can be used for combinatorial optimization problems and OR-parallel computations as well. Both depth-first and best-first search of the state space can be performed using the SearchPak. With SearchPak, a user just provides...
Learning latent space representations of high-dimensional world states has been at the core of recent rapid growth in reinforcement learning(RL). At the same time, RL algo- rithms have suffered from ignored uncertainties in the predicted estimates of model-free or model-based methods. In our work, we investigate both of these aspects...
Papers proposing novel machine learning algorithms tend to present the algorithm or technique in question in the best possible light. The standard practice is generally for authors to emphasize their proposed algorithms' performance in the precise setting where it is maximally impressive, often by only fully evaluating their best known...
We explore the application of deep learning to the disparate fields of natural language processing and computational biology. Both the sentences uttered by humans as well as the RNA and protein sequences found within the cells of their bodies can be considered formal languages in computer science, as sets of...
Assessing AI systems is difficult. Humans rely on AI systems in increasing ways, both visible and invisible, meaning a variety of stakeholders need a variety of assessment tools (e.g., a professional auditor, a developer, and an end user all have different needs). We posit that it is possible to provide...
Manufacturing technology has continuously evolved and advanced over the past century; this has led to an increase in the production of consumer and industrial goods driven by simultaneous growth in population and wealth. Despite the resulting economic and labor growth, environmental impacts of manufacturing have increased dramatically due to the...
Top-performing approaches to embodied AI tasks like point-goal navigation often rely on training agents via reinforcement learning over tens of millions (or even billions) of experiential steps – learning neural agents that map directly from visual observations to actions. In this work, we question whether these extreme training durations are...
In this work, we study the problem of learning and improving policies for probabilistic planning problems. In the first part, we train neural network policies for probabilistic planning problems modeled as factored Markov decision problems. The objective is to train problem-specific neural networks via supervised learning to imitate the action...
Causal inference is an important analytical tool to bridge the gap between prediction and decision-making. However, learning a causal network solely from data is a challenging task. In this work, various techniques have been explored for a better and improved causal network learning from data. Firstly, the problem of learning...
Advances in deep learning based image processing have led to their adoption for a wide range of applications, and in tow with these developments is a dramatic increase in the availability of high quality datasets. With this comes the need to accelerate and scale deep learning applications in order to...
Over the last two decades, satisfiability and satisfiability-modulo theory (SAT/SMT) solvers have grown powerful enough to be general purpose reasoning engines throughout software engineering and computer science. However, most practical use cases of SAT/SMT solvers require not just solving a single SAT/SMT problem, but solving sets of related SAT/SMT problems....
We propose an approach for understanding control policies represented as recurrent neural networks. Recent work has approached this problem by transforming such recurrent policy networks into finite-state machines (FSM) and then analyzing the equivalent minimized FSM. While this led to interesting insights, the minimization process can obscure a deeper understanding...
We present a novel multi-objective optimization methodology built upon a multi-agent blackboard framework. This multi-agent blackboard system (MABS) synthesizes blackboard architectures, multi-agent environments, and optimization theory. The blackboard architecture creates the framework for initializing, storing, and solving a multi-objective optimization problem. Multiple agents allow for an optimization problem to be...
Information about named entities (real-world objects) is usually harvested from different sources and organized as a multiple relational directed graph in Knowledge Bases (KBs). KBs play essential roles in many NLP modules including question answering, fact-checking, search engines, etc. KBs are big but still incomplete: relational information among entities is...
The performance of deep learning frameworks could be significantly improved through considering the particular underlying structures for each dataset. In this thesis, I summarize our three work about boosting the performance of deep learning models through leveraging structures of the data. In the first work, we theoretically justify that, for...
It is common practice in the unsupervised anomaly detection literature to create experimental benchmarks by sampling from existing supervised learning datasets. We seek to improve this practice by identifying four dimensions important to real-world anomaly detection applications --- point difficulty, clusteredness of anomalies, relevance of features, and relative frequency of...
This dissertation incorporates coalition formation and probabilistic planning towards a domain-independent automated planning solution scalable to multiple heterogeneous robots in complex domains. The first research direction investigates the effectiveness of Task Fusion and introduces heuristics that improve task allocation and result in better quality plans, while requiring lower computational cost...
Simultaneous translation, which translates concurrently with the source language speech, is widely used in many scenarios including multilateral organizations. However, it is well known to be one of the most challenging tasks for humans due to the simultaneous perception and production in two languages. On the other hand, simultaneous translation...
Diffusion processes in networks are common models for many domains, including species colonization, information/idea cascade, disease propagation and fire spreading. In diffusion networks, a diffusion event occurs when a behavior spreads from one node to the other following a probabilistic model, where the behavior could be species, an idea, a...
Natural Language Comprehension is a challenging domain of Natural Language Processing. To improve a model’s language comprehension/understanding, one approach would be to enrich the structure of the model to enhance its capability in learning the latent rules of the language.
In this dissertation, we will first introduce several deep models...
Machine learning (ML) and deep learning (DL) models impact our daily lives with applications in natural language modeling, image analysis, healthcare, genomics, and bioinformatics. The exponential growth of biological sequence data necessitates accompanying advances in computational methods. Although deep learning is highly effective for detecting and classifying biological sequences, challenges...
RNA structure prediction is a challenging problem, especially with pseudoknots. Recently, there has been a shift from the classical minimum free energy-based methods (MFE) to partition function-based ones that assemble structures based on base-pairing probabilities. Two typical examples of the latter group are the popular maximum expected accuracy (MEA) method...
Our goal is to build a system to model the RNA sequences that reveals their structural information by using efficient dynamic programming algorithms and deep learning approaches. We aim to 1) achieve linear-time for RNA secondary structure prediction based on existing minimum free energy models; 2) utilize deep neural networks...
Learning Analytics and other branches of Educational Research such as Computing Education Research (CER) implicitly assume that students, especially college students, have no barriers to access learning platforms or software packages. This assumption may be attributed to such pervasive beliefs such as "everyone has a device", or "everyone can access...
Situated cognition theory emphasizes the role that social and material contexts have on learning and knowledge application. Several studies of engineering workplace environments have noted differences between the social and material contexts of the workplace and those of undergraduate engineering education. No existing research has studied the social and material...