Physical Sciences and Mathematics Commons^™

Conditional Computation In Deep And Recurrent Neural Networks, Andrew Scott Davis

Doctoral Dissertations

Recently, deep learning models such as convolutional and recurrent neural networks have displaced state-of-the-art techniques in a variety of application domains. While the computationally heavy process of training is usually conducted on powerful graphics processing units (GPUs) distributed in large computing clusters, the resulting models can still be somewhat heavy, making deployment in resource- constrained environments potentially problematic. In this work, we build upon the idea of conditional computation, where the model is given the capability to learn how to avoid computing parts of the graph. This allows for models where the number of parameters (and in a sense, the …

Interactive Logical Analysis Of Planning Domains, Rajesh Kalyanam

Open Access Dissertations

Humans exhibit a significant ability to answer a wide range of questions about previously unencountered planning domains, and leverage this ability to construct “general-purpose'' solution plans for the domain.

The long term vision of this research is to automate this ability, constructing a system that utilizes reasoning to automatically verify claims about a planning domain. The system would use this ability to automatically construct and verify a generalized plan to solve any planning problem in the domain. The goal of this thesis is to start with baseline results from the interactive verification of claims about planning domains and develop the …

Data Driven Low-Bandwidth Intelligent Control Of A Jet Engine Combustor, Nathan L. Toner

Open Access Dissertations

This thesis introduces a low-bandwidth control architecture for navigating the input space of an un-modeled combustor system between desired operating conditions while avoiding regions of instability and blow-out. An experimental procedure is discussed for identifying regions of instability and gathering sufficient data to build a data-driven model of the system's operating modes. Regions of instability and blow-out are identified experimentally and a data-driven operating point classifier is designed. This classifier acts as a map of the operating space of the combustor, indicating regions in which the flame is in a "good" or "bad" operating mode. A data-driven predictor is also …

Robust Algorithms For Estimating Vehicle Movement From Motion Sensors Within Smartphones, Ilyas Ustun

Computational Modeling & Simulation Engineering Theses & Dissertations

Building sustainable traffic control solutions for urban streets (e.g., eco-friendly signal control) and highways requires effective and reliable sensing capabilities for monitoring traffic flow conditions so that both the temporal and spatial extents of congestion are observed. This would enable optimal control strategies to be implemented for maximizing efficiency and for minimizing the environmental impacts of traffic. Various types of traffic detection systems, such as inductive loops, radar, and cameras have been used for these purposes. However, these systems are limited, both in scope and in time. Using GPS as an alternative method is not always viable because of problems …

New Developments In Metaheuristics And Their Applications: Selected Extended Contributions From The 10th Metaheuristics International Conference (Mic 2013), Hoong Chuin Lau, Günther R. Raidl, Pascal Van Hentenryck

Research Collection School Of Computing and Information Systems

No abstract provided.

A Fast Algorithm For Personalized Travel Planning Recommendation, Aldy Gunawan, Hoong Chuin Lau, Kun Lu

Research Collection School Of Computing and Information Systems

With the pervasive use of recommender systems and web/mobile applications such as TripAdvisor and Booking.com, an emerging interest is to generate personalized tourist routes based on a tourist’s preferences and time budget constraints, often in real-time. The problem is generally known as the Tourist Trip Design Problem (TTDP) which is a route-planning problem on multiple Points of Interest (POIs). TTDP can be considered as an extension of the classical problem of Team Orienteering Problem with Time Windows (TOPTW). The objective of the TOPTW is to determine a fixed number of routes that maximize the total collected score. The TOPTW also …

Enhancing Local Search With Adaptive Operator Ordering And Its Application To The Time Dependent Orienteering Problem, Aldy Gunawan, Hoong Chuin Lau, Kun Lu

Research Collection School Of Computing and Information Systems

No abstract provided.

Artificial Intelligence And Amikacin Exposures Predictive Of Outcomes In Multidrug-Resistant Tuberculosis Patients, Chawangwa Modongo, Jotam G. Pasipanodya, Shashikant Srivastava, Nicola Zetola, Scott Williams, Giorgio Sirugo, Tawanda Gumbo

Dartmouth Scholarship

Aminoglycosides such as amikacin continue to be part of the backbone of treatment of multidrug-resistant tuberculosis (MDR- TB). We measured amikacin concentrations in 28 MDR-TB patients in Botswana receiving amikacin therapy together with oral levofloxacin, ethionamide, cycloserine, and pyrazinamide and calculated areas under the concentration-time curves from 0 to 24 h (AUC0 –24). The patients were followed monthly for sputum culture conversion based on liquid cultures. The median duration of amikacin therapy was 184 (range, 28 to 866) days, at a median dose of 17.30 (range 11.11 to 19.23) mg/kg. Only 11 (39%) pa- tients had sputum culture conversion during …

Vision-Based Motion For A Humanoid Robot, Khalid Abdullah Alkhulayfi

Dissertations and Theses

The overall objective of this thesis is to build an integrated, inexpensive, human-sized humanoid robot from scratch that looks and behaves like a human. More specifically, my goal is to build an android robot called Marie Curie robot that can act like a human actor in the Portland Cyber Theater in the play Quantum Debate with a known script of every robot behavior. In order to achieve this goal, the humanoid robot need to has degrees of freedom (DOF) similar to human DOFs. Each part of the Curie robot was built to achieve the goal of building a complete humanoid …

A Genetic Algorithmic Approach To Automated Auction Mechanism Design, Jinzhong Niu, Simon Parsons

Publications and Research

In this paper, we present a genetic algorithmic approach to automated auction mechanism design in the context of \cat games. This is a follow-up to one piece of our prior work in the domain, the reinforcement learning-based grey-box approach. Our experiments show that given the same search space the grey-box approach is able to produce better auction mechanisms than the genetic algorithmic approach. The comparison can also shed light on the design and evaluation of similar search solutions to other domain problems.

Climbing Up Cloud Nine: Performance Enhancement Techniques For Cloud Computing Environments, Mohamed Abusharkh

Electronic Thesis and Dissertation Repository

With the transformation of cloud computing technologies from an attractive trend to a business reality, the need is more pressing than ever for efficient cloud service management tools and techniques. As cloud technologies continue to mature, the service model, resource allocation methodologies, energy efficiency models and general service management schemes are not yet saturated. The burden of making this all tick perfectly falls on cloud providers. Surely, economy of scale revenues and leveraging existing infrastructure and giant workforce are there as positives, but it is far from straightforward operation from that point. Performance and service delivery will still depend on …

Constraint Answer Set Programming Versus Satisfiability Modulo Theories, Yuliya Lierler, Benjamin Susman

Computer Science Faculty Proceedings & Presentations

Constraint answer set programming is a promising research direction that integrates answer set programming with constraint processing. It is often informally related to the field of Satisfiability Modulo Theories. Yet, the exact formal link is obscured as the terminology and concepts used in these two research areas differ. In this paper, we make the link between these two areas precise.

Scalable Greedy Algorithms For Task/Resource Constrained Multi-Agent Stochastic Planning, Pritee Agrawal, Pradeep Varakantham, William Yeoh

Research Collection School Of Computing and Information Systems

Synergistic interactions between task/resource allocation and stochastic planning exist in many environments such as transportation and logistics, UAV task assignment and disaster rescue. Existing research in exploiting these synergistic interactions between the two problems have either only considered domains where tasks/resources are completely independent of each other or have focussed on approaches with limited scalability. In this paper, we address these two limitations by introducing a generic model for task/resource constrained multi-agent stochastic planning, referred to as TasC-MDPs. We provide two scalable greedy algorithms, one of which provides posterior quality guarantees. Finally, we illustrate the high scalability and solution performance …

Outlier-Robust Tensor Pca, Pan Zhou, Jiashi Feng

Research Collection School Of Computing and Information Systems

Low-rank tensor analysis is important for various real applications in computer vision. However, existing methods focus on recovering a low-rank tensor contaminated by Gaussian or gross sparse noise and hence cannot effectively handle outliers that are common in practical tensor data. To solve this issue, we propose an outlier-robust tensor principle component analysis (OR-TPCA) method for simultaneous low-rank tensor recovery and outlier detection. For intrinsically low-rank tensor observations with arbitrary outlier corruption, OR-TPCA is the first method that has provable performance guarantee for exactly recovering the tensor subspace and detecting outliers under mild conditions. Since tensor data are naturally high-dimensional …

Robust Repositioning To Counter Unpredictable Demand In Bike Sharing Systems, Supriyo Ghosh, Michael Trick, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

Bike Sharing Systems (BSSs) experience a significant loss in customer demand due to starvation (empty base stations precluding bike pickup) or congestion (full base stations precluding bike return). Therefore, BSSs operators reposition bikes between stations with the help of carrier vehicles. Due to unpredictable and dynamically changing nature of the demand, myopic reasoning typically provides a below par performance. We propose an online and robust repositioning approach to minimise the loss in customer demand while considering the possible uncertainty in future demand. Specifically, we develop a scenario generation approach based on an iterative two player game to compute a strategy …

A Hybrid Tabu/Scatter Search Algorithm For Simulation-Based Optimization Of Multi-Objective Runway Operations Scheduling, Bulent Soykan

Engineering Management & Systems Engineering Theses & Dissertations

As air traffic continues to increase, air traffic flow management is becoming more challenging to effectively and efficiently utilize airport capacity without compromising safety, environmental and economic requirements. Since runways are often the primary limiting factor in airport capacity, runway operations scheduling emerge as an important problem to be solved to alleviate flight delays and air traffic congestion while reducing unnecessary fuel consumption and negative environmental impacts. However, even a moderately sized real-life runway operations scheduling problem tends to be too complex to be solved by analytical methods, where all mathematical models for this problem belong to the complexity class …

Sequential Decision Making For Improving Efficiency In Urban Environments, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

Rapid "urbanization" (more than 50% of world's population now resides in cities) coupled with the natural lack of coordination in usage of common resources (ex: bikes, ambulances, taxis, traffic personnel, attractions) has a detrimental effect on a wide variety of response (ex: waiting times, response time for emergency needs) and coverage metrics (ex: predictability of traffic/security patrols) in cities of today. Motivated by the need to improve response and coverage metrics in urban environments, my research group is focussed on building intelligent agent systems that make sequential decisions to continuously match available supply of resources to an uncertain demand for …

Constraint Answer Set Programming Versus Satisfiability Modulo Theories, Yuliya Lierler, Benjamin Susman

Yuliya Lierler

Monte Carlo Approaches To Parameterized Poker Squares, Todd W. Neller, Zuozhi Yang, Colin M. Messinger, Calin Anton, Karo Castro-Wunsch, William Maga, Steven Bogaerts, Robert Arrington, Clay Langely

Computer Science Faculty Publications

The paper summarized a variety of Monte Carlo approaches employed in the top three performing entries to the Parameterized Poker Squares NSG Challenge competition. In all cases AI players benefited from real-time machine learning and various Monte Carlo game-tree search techniques.

Digital Integration, Jacob C. Boccio

USF Tampa Graduate Theses and Dissertations

Artificial intelligence is an emerging technology; something far beyond smartphones, cloud integration, or surgical microchip implantation. Utilizing the work of Ray Kurzweil, Nick Bostrom, and Steven Shaviro, this thesis investigates technology and artificial intelligence through the lens of the cinema. It does this by mapping contemporary concepts and the imagined worlds in film as an intersection of reality and fiction that examines issues of individual identity and alienation. I look at a non-linear timeline of films involving machine advancement, machine intelligence, and stages of post-human development; Elysium (2013) and Surrogates (2009) are about technology as an extension of the self, …

Energy Consumption Prediction With Big Data: Balancing Prediction Accuracy And Computational Resources, Katarina Grolinger, Miriam Am Capretz, Luke Seewald

Electrical and Computer Engineering Publications

In recent years, advances in sensor technologies and expansion of smart meters have resulted in massive growth of energy data sets. These Big Data have created new opportunities for energy prediction, but at the same time, they impose new challenges for traditional technologies. On the other hand, new approaches for handling and processing these Big Data have emerged, such as MapReduce, Spark, Storm, and Oxdata H2O. This paper explores how findings from machine learning with Big Data can benefit energy consumption prediction. An approach based on local learning with support vector regression (SVR) is presented. Although local learning itself is …

Analysis On Alergia Algorithm: Pattern Recognition By Automata Theory, Xuanyi Qi

Master's Projects

Based on Kolmogorov Complexity, a finite set x of strings has a pattern if the set x can be output by a Turing machine of length that is less than minimum of all |x|; this Turing machine, that may not be unique, is called a pattern of the finite set of string. In order to find a pattern of a given finite set of strings (assuming such a pattern exists), the ALERGIA algorithm is used to approximate such a pattern (Turing machine) in terms of finite automata. Note that each finite automaton defines a partition on formal language Σ*, ALERGIA …

Analyze Large Multidimensional Datasets Using Algebraic Topology, David Le

Master's Projects

This paper presents an efficient algorithm to extract knowledge from high-dimensionality, high- complexity datasets using algebraic topology, namely simplicial complexes. Based on concept of isomorphism of relations, our method turn a relational table into a geometric object (a simplicial complex is a polyhedron). So, conceptually association rule searching is turned into a geometric traversal problem. By leveraging on the core concepts behind Simplicial Complex, we use a new technique (in computer science) that improves the performance over existing methods and uses far less memory. It was designed and developed with a strong emphasis on scalability, reliability, and extensibility. This paper …

Dna Analysis Using Grammatical Inference, Cory Cook

Master's Projects

An accurate language definition capable of distinguishing between coding and non-coding DNA has important applications and analytical significance to the field of computational biology. The method proposed here uses positive sample grammatical inference and statistical information to infer languages for coding DNA.

An algorithm is proposed for the searching of an optimal subset of input sequences for the inference of regular grammars by optimizing a relevant accuracy metric. The algorithm does not guarantee the finding of the optimal subset; however, testing shows improvement in accuracy and performance over the basis algorithm.

Testing shows that the accuracy of inferred languages for …

Optimizing The Mix Of Games And Their Locations On The Casino Floor, Jason D. Fiege, Anastasia D. Baran

International Conference on Gambling & Risk Taking

We present a mathematical framework and computational approach that aims to optimize the mix and locations of slot machine types and denominations, plus other games to maximize the overall performance of the gaming floor. This problem belongs to a larger class of spatial resource optimization problems, concerned with optimizing the allocation and spatial distribution of finite resources, subject to various constraints. We introduce a powerful multi-objective evolutionary optimization and data-modelling platform, developed by the presenter since 2002, and show how this software can be used for casino floor optimization. We begin by extending a linear formulation of the casino floor …

Stationary And Time-Dependent Optimization Of The Casino Floor Slot Machine Mix, Anastasia D. Baran, Jason D. Fiege

International Conference on Gambling & Risk Taking

Modeling and optimizing the performance of a mix of slot machines on a gaming floor can be addressed at various levels of coarseness, and may or may not consider time-dependent trends. For example, a model might consider only time-averaged, aggregate data for all machines of a given type; time-dependent aggregate data; time-averaged data for individual machines; or fully time dependent data for individual machines. Fine-grained, time-dependent data for individual machines offers the most potential for detailed analysis and improvements to the casino floor performance, but also suffers the greatest amount of statistical noise. We present a theoretical analysis of single …

Machine Learning On The Cloud For Pattern Recognition, Tien Nguyen

Master's Projects

Pattern recognition is a field of machine learning with applications to areas such as text recognition and computer vision. Machine learning algorithms, such as convolutional neural networks, may be trained to classify images. However, such tasks may be computationally intensive for a commercial computer for larger volumes or larger sizes of images. Cloud computing allows one to overcome the processing and memory constraints of average commercial computers, allowing computations on larger amounts of data. In this project, we developed a system for detection and tracking of moving human and vehicle objects in videos in real time or near real time. …

Data-Driven Synthesis And Evaluation Of Syntactic Facial Expressions In American Sign Language Animation, Hernisa Kacorri

Dissertations, Theses, and Capstone Projects

Technology to automatically synthesize linguistically accurate and natural-looking animations of American Sign Language (ASL) would make it easier to add ASL content to websites and media, thereby increasing information accessibility for many people who are deaf and have low English literacy skills. State-of-art sign language animation tools focus mostly on accuracy of manual signs rather than on the facial expressions. We are investigating the synthesis of syntactic ASL facial expressions, which are grammatically required and essential to the meaning of sentences. In this thesis, we propose to: (1) explore the methodological aspects of evaluating sign language animations with facial expressions, …

Approaching Humans For Help: A Study Of Human-Robot Proxemics, Eric Rose

Honors Theses

In order for a robot to be effective when interacting with a person, it is important for the robot to choose the correct person. Consider an example where a robot is trying to perform a task but it isn’t capable of doing a subtask, like going up a flight of stairs. In this case, the robot would need to ask a person for help with the elevator, in a socially appropriate way. We have conducted an experiment to determine who would be the best candidate to approach in a situation like this. Should the robot choose to approach someone who …

Self-Organizing Neural Network For Adaptive Operator Selection In Evolutionary Search, Teck Hou Teng, Stephanus Daniel Handoko, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

Evolutionary Algorithm is a well-known meta-heuristics paradigm capable of providing high-quality solutions to computationally hard problems. As with the other meta-heuristics, its performance is often attributed to appropriate design choices such as the choice of crossover operators and some other parameters. In this chapter, we propose a continuous state Markov Decision Process model to select crossover operators based on the states during evolutionary search. We propose to find the operator selection policy efficiently using a self-organizing neural network, which is trained offline using randomly selected training samples. The trained neural network is then verified on test instances not used for …