Physical Sciences and Mathematics Commons^™

Limitations And Extensions Of The Wolf-Phc Algorithm, Philip R. Cook

Theses and Dissertations

Policy Hill Climbing (PHC) is a reinforcement learning algorithm that extends Q-learning to learn probabilistic policies for multi-agent games. WoLF-PHC extends PHC with the "win or learn fast" principle. A proof that PHC will diverge in self-play when playing Shapley's game is given, and WoLF-PHC is shown empirically to diverge as well. Various WoLF-PHC based modifications were created, evaluated, and compared in an attempt to obtain convergence to the single shot Nash equilibrium when playing Shapley's game in self-play without using more information than WoLF-PHC uses. Partial Commitment WoLF-PHC (PCWoLF-PHC), which performs best on Shapley's game, is tested on other …

Improving Neural Network Classification Training, Michael Edwin Rimer

Theses and Dissertations

The following work presents a new set of general methods for improving neural network accuracy on classification tasks, grouped under the label of classification-based methods. The central theme of these approaches is to provide problem representations and error functions that more directly improve classification accuracy than conventional learning and error functions. The CB1 algorithm attempts to maximize classification accuracy by selectively backpropagating error only on misclassified training patterns. CB2 incorporates a sliding error threshold to the CB1 algorithm, interpolating between the behavior of CB1 and standard error backpropagation as training progresses in order to avoid prematurely saturated network weights. CB3 …

Parallelization Of Ant Colony Optimization Via Area Of Expertise Learning, Adrian A. De Freitas

Theses and Dissertations

Ant colony optimization algorithms have long been touted as providing an effective and efficient means of generating high quality solutions to NP-hard optimization problems. Unfortunately, while the structure of the algorithm is easy to parallelize, the nature and amount of communication required for parallel execution has meant that parallel implementations developed suffer from decreased solution quality, slower runtime performance, or both. This thesis explores a new strategy for ant colony parallelization that involves Area of Expertise (AOE) learning. The AOE concept is based on the idea that individual agents tend to gain knowledge of different areas of the search space …

Solar Activity Detection And Prediction Using Image Processing And Machine Learning Techniques, Gang Fu

Dissertations

The objective of the research in this dissertation is to develop the methods for automatic detection and prediction of solar activities, including prominence eruptions, emerging flux regions and solar flares. Image processing and machine learning techniques are applied in this study. These methods can be used for automatic observation of solar activities and prediction of space weather that may have great influence on the near earth environment.

The research presented in this dissertation covers the following topics: i) automatic detection of prominence eruptions (PBs), ii) automatic detection of emerging flux regions (EFRs), and iii) automatic prediction of solar flares.

In …

Predicting Coronary Artery Disease With Medical Profile And Gene Polymorphisms Data, Qiongyu Chen, Guoliang Li, Tze-Yun Leong, Chew-Kiat Heng

Research Collection School Of Computing and Information Systems

Coronary artery disease (CAD) is a main cause of death in the world. Finding cost-effective methods to predict CAD is a major challenge in public health. In this paper, we investigate the combined effects of genetic polymorphisms and non-genetic factors on predicting the risk of CAD by applying well known classification methods, such as Bayesian networks, naïve Bayes, support vector machine, k-nearest neighbor, neural networks and decision trees. Our experiments show that all these classifiers are comparable in terms of accuracy, while Bayesian networks have the additional advantage of being able to provide insights into the relationships among the variables. …

Obstacle Avoidance And Path Traversal Using Interactive Machine Learning, Jonathan M. Turner

Theses and Dissertations

Recently there has been a growing interest in using robots in activities that are dangerous or cost prohibitive for humans to do. Such activities include military uses and space exploration. While robotic hardware is often capable of being used in these types of situations, the ability of human operators to control robots in an effective manner is often limited. This deficiency is often related to the control interface of the robot and the level of autonomy that control system affords the human operator. This thesis describes a robot control system, called the safe/unsafe system, which gives a human operator the …

Cognitive And Behavioral Model Ensembles For Autonomous Virtual Characters, Jeffrey S. Whiting

Theses and Dissertations

Cognitive and behavioral models have become popular methods to create autonomous self-animating characters. Creating these models presents the following challenges: (1) Creating a cognitive or behavioral model is a time intensive and complex process that must be done by an expert programmer (2) The models are created to solve a specific problem in a given environment and because of their specific nature cannot be easily reused. Combining existing models together would allow an animator, without the need of a programmer, to create new characters in less time and would be able to leverage each model's strengths to increase the character's …

Active Learning For Part-Of-Speech Tagging: Accelerating Corpus Annotation, George Busby, Marc Carmen, James Carroll, Robbie Haertel, Deryle W. Lonsdale, Peter Mcclanahan, Eric K. Ringger, Kevin Seppi

Faculty Publications

In the construction of a part-of-speech annotated corpus, we are constrained by a fixed budget. A fully annotated corpus is required, but we can afford to label only a subset. We train a Maximum Entropy Markov Model tagger from a labeled subset and automatically tag the remainder. This paper addresses the question of where to focus our manual tagging efforts in order to deliver an annotation of highest quality. In this context, we find that active learning is always helpful. We focus on Query by Uncertainty (QBU) and Query by Committee (QBC) and report on experiments with several baselines and …

Evolutionary Granular Kernel Machines, Bo Jin

Computer Science Dissertations

Kernel machines such as Support Vector Machines (SVMs) have been widely used in various data mining applications with good generalization properties. Performance of SVMs for solving nonlinear problems is highly affected by kernel functions. The complexity of SVMs training is mainly related to the size of a training dataset. How to design a powerful kernel, how to speed up SVMs training and how to train SVMs with millions of examples are still challenging problems in the SVMs research. For these important problems, powerful and flexible kernel trees called Evolutionary Granular Kernel Trees (EGKTs) are designed to incorporate prior domain knowledge. …

Learning To Classify E-Mail, Irena Koprinska, Josiah Poon, James Clark, Jason Yuk Hin Chan

Research Collection School Of Computing and Information Systems

In this paper we study supervised and semi-supervised classification of e-mails. We consider two tasks: filing e-mails into folders and spam e-mail filtering. Firstly, in a supervised learning setting, we investigate the use of random forest for automatic e-mail filing into folders and spam e-mail filtering. We show that random forest is a good choice for these tasks as it runs fast on large and high dimensional databases, is easy to tune and is highly accurate, outperforming popular algorithms such as decision trees, support vector machines and naive Bayes. We introduce a new accurate feature selector with linear time complexity. …

Towards A Self-Calibrating Video Camera Network For Content Analysis And Forensics, Imran Junejo

Electronic Theses and Dissertations

Due to growing security concerns, video surveillance and monitoring has received an immense attention from both federal agencies and private firms. The main concern is that a single camera, even if allowed to rotate or translate, is not sufficient to cover a large area for video surveillance. A more general solution with wide range of applications is to allow the deployed cameras to have a non-overlapping field of view (FoV) and to, if possible, allow these cameras to move freely in 3D space. This thesis addresses the issue of how cameras in such a network can be calibrated and how …

Using Machine Learning Techniques To Create Ai Controlled Players For Video Games, Bhuman Soni

Theses : Honours

This study aims to achieve higher replay and entertainment value in a game through human-like AI behaviour in computer controlled characters called bats. In order to achieve that, an artificial intelligence system capable of learning from observation of human player play was developed. The artificial intelligence system makes use of machine learning capabilities to control the state change mechanism of the bot. The implemented system was tested by an audience of gamers and compared against bats controlled by static scripts. The data collected was focused on qualitative aspects of replay and entertainment value of the game and subjected to quantitative …

Knowledge-Based Methods For Automatic Extraction Of Domain-Specific Ontologies, Janardhana R. Punuru

LSU Doctoral Dissertations

Semantic web technology aims at developing methodologies for representing large amount of knowledge in web accessible form. The semantics of knowledge should be easy to interpret and understand by computer programs, so that sharing and utilizing knowledge across the Web would be possible. Domain specific ontologies form the basis for knowledge representation in the semantic web. Research on automated development of ontologies from texts has become increasingly important because manual construction of ontologies is labor intensive and costly, and, at the same time, large amount of texts for individual domains is already available in electronic form. However, automatic extraction of …

Learning In Short-Time Horizons With Measurable Costs, Patrick Bowen Mullen

Theses and Dissertations

Dynamic pricing is a difficult problem for machine learning. The environment is noisy, dynamic and has a measurable cost associated with exploration that necessitates that learning be done in short-time horizons. These short-time horizons force the learning algorithms to make pricing decisions based on scarce data. In this work, various machine learning algorithms are compared in the context of dynamic pricing. These algorithms include the Kalman filter, artificial neural networks, particle swarm optimization and genetic algorithms. The majority of these algorithms have been modified to handle the pricing problem. The results show that these adaptations allow the learning algorithms to …

A Decentralized Reinforcement Learning Controller For Collaborative Driving, Luke Ng, Christopher M. Clark, Jan P. Huissoon

Computer Science and Software Engineering

Research in the collaborative driving domain strives to create control systems that coordinate the motion of multiple vehicles in order to navigate traffic both efficiently and safely. In this paper a novel individual vehicle controller based on reinforcement learning is introduced. This controller is capable of both lateral and longitudinal control while driving in a multi-vehicle platoon. The design and development of this controller is discussed in detail and simulation results showing learning progress and performance are presented.

Svm-Based Negative Data Mining To Binary Classification, Fuhua Jiang

Computer Science Dissertations

The properties of training data set such as size, distribution and the number of attributes significantly contribute to the generalization error of a learning machine. A not well-distributed data set is prone to lead to a partial overfitting model. Two approaches proposed in this dissertation for the binary classification enhance useful data information by mining negative data. First, an error driven compensating hypothesis approach is based on Support Vector Machines (SVMs) with (1+k)-iteration learning, where the base learning hypothesis is iteratively compensated k times. This approach produces a new hypothesis on the new data set in which each label is …

Particle Swarm Optimization In Dynamic Pricing, Christopher K. Monson, Patrick B. Mullen, Kevin Seppi, Sean C. Warnick

Faculty Publications

Dynamic pricing is a real-time machine learning problem with scarce prior data and a concrete learning cost. While the Kalman Filter can be employed to track hidden demand parameters and extensions to it can facilitate exploration for faster learning, the exploratory nature of Particle Swarm Optimization makes it a natural choice for the dynamic pricing problem. We compare both the Kalman Filter and existing particle swarm adaptations for dynamic and/or noisy environments with a novel approach that time-decays each particle's previous best value; this new strategy provides more graceful and effective transitions between exploitation and exploration, a necessity in the …

Temporal Data Mining In A Dynamic Feature Space, Brent K. Wenerstrom

Theses and Dissertations

Many interesting real-world applications for temporal data mining are hindered by concept drift. One particular form of concept drift is characterized by changes to the underlying feature space. Seemingly little has been done to address this issue. This thesis presents FAE, an incremental ensemble approach to mining data subject to concept drift. FAE achieves better accuracies over four large datasets when compared with a similar incremental learning algorithm.

Learning Real-World Problems By Finding Correlated Basis Functions, Adam C. Drake

Theses and Dissertations

Learning algorithms based on the Fourier transform attempt to learn functions by approximating the largest coefficients of their Fourier representations. Nearly all previous work in Fourier-based learning has been in the theoretical realm, where properties of the transform have made it possible to prove many interesting learnability results. The real-world usefulness of Fourier-based methods, however, has not been thoroughly explored. This thesis explores methods for the practical application of Fourier-based learning. The primary contribution of this thesis is a new search algorithm for finding the largest coefficients of a function's Fourier representation. Although the search space is exponentially large, empirical …

Surface Realization Using A Featurized Syntactic Statistical Language Model, Thomas L. Packer

Theses and Dissertations

An important challenge in natural language surface realization is the generation of grammatical sentences from incomplete sentence plans. Realization can be broken into a two-stage process consisting of an over-generating rule-based module followed by a ranker that outputs the most probable candidate sentence based on a statistical language model. Thus far, an n-gram language model has been evaluated in this context. More sophisticated syntactic knowledge is expected to improve such a ranker. In this thesis, a new language model based on featurized functional dependency syntax was developed and evaluated. Generation accuracies and cross-entropy for the new language model did not …

K X N Trust-Based Agent Reputation, Christopher Alonzo Parker

Theses and Dissertations

In this research, a multi-agent system called KMAS is presented that models an environment of intelligent, autonomous, rational, and adaptive agents that reason about trust, and adapt trust based on experience. Agents reason and adapt using a modification of the k-Nearest Neighbor algorithm called (k X n) Nearest Neighbor where k neighbors recommend reputation values for trust during each of n interactions. Reputation allows a single agent to receive recommendations about the trustworthiness of others. One goal is to present a recommendation model of trust that outperforms MAS architectures relying solely on direct agent interaction. A second goal is to …

Task Similarity Measures For Transfer In Reinforcement Learning Task Libraries, James Carroll, Kevin Seppi

Faculty Publications

Recent research in task transfer and task clustering has necessitated the need for task similarity measures in reinforcement learning. Determining task similarity is necessary for selective transfer where only information from relevant tasks and portions of a task are transferred. Which task similarity measure to use is not immediately obvious. It can be shown that no single task similarity measure is uniformly superior. The optimal task similarity measure is dependent upon the task transfer method being employed. We define similarity in terms of tasks, and propose several possible task similarity measures, dT, dp, dQ, and dR which are based on …

Dynamically Optimized Context In Recommender Systems, Ghim-Eng Yap, Ah-Hwee Tan, Hwee Hwa Pang

Research Collection School Of Computing and Information Systems

Traditional approaches to recommender systems have not taken into account situational information when making recommendations, and this seriously limits the relevance of the results. This paper advocates context-awareness as a promising approach to enhance the performance of recommenders, and introduces a mechanism to realize this approach. We present a framework that separates the contextual concerns from the actual recommendation module, so that contexts can be readily shared across applications. More importantly, we devise a learning algorithm to dynamically identify the optimal set of contexts for a specific recommendation task and user. An extensive series of experiments has validated that our …

Improving And Extending Behavioral Animation Through Machine Learning, Jonathan J. Dinerstein

Theses and Dissertations

Behavioral animation has become popular for creating virtual characters that are autonomous agents and thus self-animating. This is useful for lessening the workload of human animators, populating virtual environments with interactive agents, etc. Unfortunately, current behavioral animation techniques suffer from three key problems: (1) deliberative behavioral models (i.e., cognitive models) are slow to execute; (2) interactive virtual characters cannot adapt online due to interaction with a human user; (3) programming of behavioral models is a difficult and time-intensive process. This dissertation presents a collection of papers that seek to overcome each of these problems. Specifically, these issues are alleviated …

Evaluating Online Trust Using Machine Learning Methods, Weihua Song

Doctoral Dissertations

Trust plays an important role in e-commerce, P2P networks, and information filtering. Current challenges in trust evaluations include: (1) fnding trustworthy recommenders, (2) aggregating heterogeneous trust recommendations of different trust standards based on correlated observations and different evaluation processes, and (3) managing efficiently large trust systems where users may be sparsely connected and have multiple local reputations. The purpose of this dissertation is to provide solutions to these three challenges by applying ordered depth-first search, neural network, and hidden Markov model techniques. It designs an opinion filtered recommendation trust model to derive personal trust from heterogeneous recommendations; develops a reputation …

Learning Discrete Hidden Markov Models From State Distribution Vectors, Luis G. Moscovich

LSU Doctoral Dissertations

Hidden Markov Models (HMMs) are probabilistic models that have been widely applied to a number of fields since their inception in the late 1960’s. Computational Biology, Image Processing, and Signal Processing, are but a few of the application areas of HMMs. In this dissertation, we develop several new efficient learning algorithms for learning HMM parameters. First, we propose a new polynomial-time algorithm for supervised learning of the parameters of a first order HMM from a state probability distribution (SD) oracle. The SD oracle provides the learner with the state distribution vector corresponding to a query string. We prove the correctness …

Collective Multi-Label Classification, Nadia Ghamrawi, Andrew Mccallum

Computer Science Department Faculty Publication Series

Common approaches to multi-label classification learn independent classifiers for each category, and employ ranking or thresholding schemes for classification. Because they do not exploit dependencies between labels, such techniques are only well-suited to problems in which categories are independent. However, in many domains labels are highly interdependent. This paper explores multilabel conditional random field (CRF) classification models that directly parameterize label co-occurrences in multi-label classification. Experiments show that the models outperform their singlelabel counterparts on standard text corpora. Even when multilabels are sparse, the models improve subset classification error by as much as 40%.

An Assessment Of Case-Based Reasoning For Spam Filtering, Sarah Jane Delany, Padraig Cunningham, Lorcan Coyle

Articles

Because of the changing nature of spam, a spam filtering system that uses machine learning will need to be dynamic. This suggests that a case-based (memory-based) approach may work well. Case-Based Reasoning (CBR) is a lazy approach to machine learning where induction is delayed to run time. This means that the case base can be updated continuously and new training data is immediately available to the induction process. In this paper we present a detailed description of such a system called ECUE and evaluate design decisions concerning the case representation. We compare its performance with an alternative system that uses …

A Bayesian Technique For Task Localization In Multiple Goal Markov Decision Processes, James Carroll, Kevin Seppi

Faculty Publications

In a reinforcement learning task library system for Multiple Goal Markov Decision Process (MGMDP), localization in the task space allows the agent to determine whether a given task is already in its library in order to exploit previously learned experience. Task localization in MGMDPs can be accomplished through a Bayesian approach, however a trivial approach fails when the rewards are not distributed normally. This can be overcome through our Bayesian Task Localization Technique (BTLT).

Vision-Based Human Directed Robot Guidance, Richard B. Arthur

Theses and Dissertations

This paper describes methods to track a user-defined point in the vision of a robot as it drives forward. This tracking allows a robot to keep itself directed at that point while driving so that it can get to that user-defined point. I develop and present two new multi-scale algorithms for tracking arbitrary points between two frames of video, as well as through a video sequence. The multi-scale algorithms do not use the traditional pyramid image, but instead use a data structure called an integral image (also known as a summed area table). The first algorithm uses edge-detection to track …