<p> While robotics has made considerable strides toward more robust and adaptive manipulation, perception, and planning, robots in the near future are unlikely to be as dexterous, competent, and versatile as human workers. Rather than try to create fully autonomous systems that accomplish tasks independently, a more practical approach is to construct robots that work alongside people. This allows human and robot workers to concentrate on the tasks for which they are each best suited, while simultaneously providing the capability to assist each other during tasks that one worker lacks the ability to complete independently in a safe or maximally proficient manner. Human-robot teaming advances have the potential to extend applications of autonomous robots well beyond their current, limited roles in factory automation settings. Much of modern robotics remains inapplicable in many domains where tasks are either too complex, beyond modern hardware limitations, too sensitive for non-human completion, or too flexible for static automation practices. In these situations human-robot teaming can be leveraged to improve the efficiency, quality-of-life, and safety of human partners.</p><p> In this thesis, I describe algorithms that can create collaborative robots that call provide assistance when useful, remove dull or undesirable responsibilities when possible, and assist with dangerous tasks when feasible. In doing so, I present a novel method for autonomously constructing hierarchical task networks that factor complex tasks in was that make theism approachable by modern planning and coordination algorithms. In particular, within these complex cooperative tasks I focus on facilitating collaboration between a lead worker and robotic assistant within a shared space, defining and investigating a class of actions I term supportive behaviors: actions that serve to reduce the cognitive or kinematic complexity of tasks for teammates. The majority of contributions within this work center around discovering, learning, and executing these types of behaviors in multi-agent domains with asymmetric authority. I provide an examination of supportive behavior learning and execution from the perspective of task and motion planning, as well as that of learning directly from interactions with humans. These algorithms provide a collaborative robot with the capability to anticipate the needs of a human teammate and proactively offer help as needed or desired. This work enables to creation of robots that provide tools just-in-time, robots that alter workspaces to make more optimal task orderings more obvious and more feasible, and robots that recognize when a user is delayed in a complex task and offer assistance.</p><p> Combining these algorithms provides a basis for a robot with both a capacity for rich task comprehension and a theory of mind about its collaborators, enabling methods to allow such a robot to leverage knowledge it acquires to transition between the role of learner, able assistant, and informative instructor during interactions with teammates.</p>

<p> Many important modern engineering problems involve satisfying multiple objectives. Simultaneous optimization of these objectives can be difficult as they compete for the same set of any given resources. One way to solve multiple-objective optimization is with the use of genetic algorithms (GA&rsquo;s). </p><p> One can break down the structure of these multi-objective genetic algorithms (MOGA&rsquo;s) into two different approaches. One approach is based on incorporating multiple objectives into a single fitness function which will evaluate how well a given solution solves the issue. The other approach uses multiple fitness functions, each representing a different objective, which when combined create a solution set of possible solutions to the problem. This project focuses on combining these approaches in order to make a hybrid model, which can benefit from combining the results of the previous two methods; incorporating a level of automation that allows for inference of a final solution based on different prioritization of each objective. This solution would not have been previously attainable by either standalone method. </p><p> This project is named the Automated Inferential Multi-Objective Optimization System (AIMOS), and it can be applied to a multitude of different problem types. In order to show its capabilities, AIMOS has been applied to a theoretical optimization problem used to measure the effectiveness of GA&rsquo;s. </p>

<p> Investigated into and motivated by Ensemble Machine Learning (<i>ML</i>) techniques, this thesis contributes to addressing performance, consistency, and integrity issues such as overfitting, underfitting, predictive errors, accuracy paradox, and poor generalization for the <i>ML</i> models. Ensemble <i>ML</i> methods have shown promising outcome when a single algorithm failed to approximate the true prediction function. Using meta-learning, a super learner is engineered by combining weak learners. Generally, several methods in Supervised Learning (<i>SL</i>) are evaluated to find the best fit to the underlying data and predictive analytics (<i>i.e.</i>, &ldquo;<i>No Free Lunch</i>&rdquo; Theorem relevance). This thesis addresses three main challenges/problems, <i> i</i>) determining the optimum blend of algorithms/methods for enhanced <i> SL</i> ensemble models, <i>ii</i>) engineering the selection and grouping of features that aggregate to the highest possible predictive and non-redundant value in the training data set, and <i>iii</i>) addressing the performance integrity issues such as accuracy paradox. Therefore, an enhanced Machine Learning Engine Engineering (<i>eMLEE</i>) is inimitably constructed via built-in parallel processing and specially designed novel constructs for error and gain functions to optimally score the classifier elements for improved training experience and validation procedures. <i> eMLEE</i>, as based on stochastic thinking, is built on; <i>i</i>) one centralized unit as Logical Table unit (<i>LT</i>), <i> ii</i>) two explicit units as enhanced Algorithm Blend and Tuning (<i> eABT</i>) and enhanced Feature Engineering and Selection (<i>eFES </i>), and two implicit constructs as enhanced Weighted Performance Metric (<i>eWPM</i>) and enhanced Cross Validation and Split (<i> eCVS</i>). Hence, it proposes an enhancement to the internals of the <i> SL</i> ensemble approaches. </p><p> Motivated by nature inspired metaheuristics algorithms (such as <i> GA, PSO, ACO</i>, etc.), feedback mechanisms are improved by introducing a specialized function as <i>Learning from the Mistakes</i> (<i> LFM</i>) to mimic the human learning experience. <i>LFM</i> has shown significant improvement towards refining the predictive accuracy on the testing data by utilizing the computational processing of wrong predictions to increase the weighting scoring of the weak classifiers and features. <i> LFM</i> further ensures the training layer experiences maximum mistakes (<i>i.e.</i>, errors) for optimum tuning. With this designed in the engine, stochastic modeling/thinking is implicitly implemented. </p><p> Motivated by OOP paradigm in the high-level programming, <i>eMLEE </i> provides interface infrastructure using <i>LT</i> objects for the main units (<i>i.e.</i>, Unit A and Unit B) to use the functions on demand during the classifier learning process. This approach also assists the utilization of <i>eMLEE</i> API by the outer real-world usage for predictive modeling to further customize the classifier learning process and tuning elements trade-off, subject to the data type and end model in goal. </p><p> Motivated by higher dimensional processing and Analysis (<i>i.e. </i>, <i>3D</i>) for improved analytics and learning mechanics, <i> eMLEE</i> incorporates <i>3D</i> Modeling of fitness metrics such as <i>x</i> for overfit, <i>y</i> for underfit, and <i> z</i> for optimum fit, and then creates logical cubes using <i> LT</i> handles to locate the optimum space during ensemble process. This approach ensures the fine tuning of ensemble learning process with improved accuracy metric. </p><p> To support the built and implementation of the proposed scheme, mathematical models (<i>i.e.</i>, <i>Definitions, Lemmas, Rules</i>, and <i>Procedures</i>) along with the governing algorithms&rsquo; definitions (and <i>pseudo-code</i>), and necessary illustrations (<i>to assist in elaborating the concepts</i>) are provided. Diverse sets of data are used to improve the generalization of the engine and tune the underlying constructs during development-testing phases. To show the practicality and stability of the proposed scheme, several results are presented with a comprehensive analysis of the outcomes for the metrics (<i>i.e.</i>, <i> via integrity, corroboration</i>, and <i>quantification</i>) of the engine. Two approaches are followed to corroborate the engine, <i> i</i>) testing inner layers (<i>i.e.</i>, internal constructs) of the engine (<i>i.e.</i>, <i>Unit-A, Unit-B</i>, and <i> C-Unit</i>) to stabilize and test the fundamentals, and <i>ii</i>) testing outer layer (<i>i.e.</i>, <i>engine as a black box </i>) for standard measuring metrics for the real-world endorsement. Comparison with various existing techniques in the state of the art are also reported. In conclusion of the extensive literature review, research undertaken, investigative approach, engine construction and tuning, validation approach, experimental study, and results visualization, the <i>eMLEE</i> is found to be outperforming the existing techniques most of the time, in terms of the classifier learning, generalization, metrics trade-off, optimum-fitness, feature engineering, and validation.</p><p>

<p> The prediction of internal human states&mdash;what a person is thinking and feeling&mdash;by computers remains an elusive yet attractive goal. The potential applications of such a system are many, including but not limited to lie detection, consumer satisfaction and interest, autonomous vehicle safety, enhanced social networking, and even robotic friendship and companionship. </p><p> We believe that scientific progress towards such lofty and exciting applications as described above has been slowed because prior efforts to create an autonomous computer system capable of predicting what a human being is thinking or feeling have been largely predicting or detecting &ldquo;emotions&rdquo; in humans, rather than detecting more empirically falsifiable events. There is much scholarly disagreement as to the exact nature of what emotions actually are, and thus Affective Computing dealing with emotional prediction is forced to rely on poorly-defined and often-contradictory ground truths. </p><p> In our research, we focus on predicting discrete events based on the expressive behavior of human subjects. Our systems use a user-dependent method of analysis and rely heavily on contextual information to make predictions about the meaning behind subject expressive behavior. Our system&rsquo;s accuracy and therefore usefulness are built on provable ground truths that prohibit the drawing of inaccurate conclusions that other systems could too easily make. </p><p> While the current application of our research idea&mdash;a Contextually Informed Program for Hyper-personalized Empathetic Report, or CIPHER&mdash;is modest (predicting player behavior during a game of poker), we believe it to be the foundation upon which much more extensive empathetic and affective analysis systems could be solidly and eventually built.</p><p>

<p> To create algorithmic art with documents available on the internet, artists must discover strategies for organizing those documents. In this project I used a graph structure based on Melissa Schilling's model of cognitive insight to reorganize sounds on the web using aural and lexical relationships. I was then able to generate music with these graphs using several different activation strategies. In section one I introduce my goals for this project. In section two I review other approaches to this problem and art that has influenced my approach. In section three I demonstrate techniques for organizing and collaging sounds from freesound.org. Sounds can be organized in a graph structure by exploiting aural similarity relationships provided by freesound.org, and lexical relationships provided by wordnik.com. Music can then be generated from these graphs in a variety of ways. In section four I show how my software was inspired by theories of creativity. Specifically I show how my software is an illustration of Melissa Schilling's graph model of cognitive insight. In section five, I elaborate on the pieces I've generated for this dissertation using this software and several other novel sound generating programs.</p>

<p> Our daily lives increasingly involve interactions with others via different communication channels, such as email, text messaging, and social media. In this context, the ability to analyze and understand our communication patterns is becoming increasingly important. This dissertation focuses on generative probabilistic models for describing different characteristics of communication behavior, focusing primarily on email communication. </p><p> First, we present a two-parameter kernel density estimator for estimating the probability density over recipients of an email (or, more generally, items which appear in an itemset). A stochastic gradient method is proposed for efficiently inferring the kernel parameters given a continuous stream of data. Next, we apply the kernel model and the Bernoulli mixture model to two important prediction tasks: given a partially completed email recipient list, 1) predict which others will be included in the email, and 2) rank potential recipients based on their likelihood to be added to the email. Such predictions are useful in suggesting future actions to the user (i.e. which person to add to an email) based on their previous actions. We then investigate a piecewise-constant Poisson process model for describing the time-varying communication rate between an individual and several groups of their contacts, where changes in the Poisson rate are modeled as latent state changes within a hidden Markov model. </p><p> We next focus on the time it takes for an individual to respond to an event, such as receiving an email. We show that this response time depends heavily on the individual's typical daily and weekly patterns - patterns not adequately captured in standard models of response time (e.g. the Gamma distribution or Hawkes processes). A time-warping mechanism is introduced where the absolute response time is modeled as a transformation of effective response time, relative to the daily and weekly patterns of the individual. The usefulness of applying the time-warping mechanism to standard models of response time, both in terms of log-likelihood and accuracy in predicting which events will be quickly responded to, is illustrated over several individual email histories.</p>

Thesis (M.S.)--Youngstown State University, 2009. / Includes bibliographical references (leaves 51-52). Also available via the World Wide Web in PDF format.

Thesis (Ph.D.)--Indiana University, Dept. of Mathematics and Cognitive Science, 2009. / Title from PDF t.p. (viewed on Jul 19, 2010). Source: Dissertation Abstracts International, Volume: 70-12, Section: B, page: 7604. Adviser: Lawrence S. Moss.

<p> This paper introduces a novel approach for assessing multiple patterns in biological imaging datasets. The developed tool should be able to provide most probable structure of a dataset of images that consists of biological patterns not encountered during the model training process. The tool includes two major parts: (1) feature learning and extraction pipeline and (2) subsequent clustering with estimation of number of classes. The feature-learning part includes two deep-learning techniques and a feature quantitation pipeline as a benchmark method. Clustering includes three non-parametric methods. K-means clustering is employed for validation and hypothesis testing by comparing results with provided ground truth. The most appropriate methods and hyper-parameters were suggested to achieve maximum clustering quality. A convolutional autoencoder demonstrated the most stable and robust results: entropy-based V-measure metric 0.9759 on a dataset of classes employed for training and 0.9553 on a dataset of completely novel classes.</p><p>

<p> Markov random walk models are powerful analytical tools for multiple areas in machine learning, numerical optimizations and data mining tasks. The key assumption of a first-order Markov chain is memorylessness, which restricts the dependence of the transition distribution to the current state only. However in many applications, this assumption is not appropriate. We propose a set of higher-order random walk techniques and discuss their applications to tensor co-clustering, user trails modeling, and solving linear systems. First, we develop a new random walk model that we call the super-spacey random surfer, which simultaneously clusters the rows, columns, and slices of a nonnegative three-mode tensor. This algorithm generalizes to tensors with any number of modes. We partition the tensor by minimizing the exit probability between clusters when the super-spacey random walk is at stationary. The second application is user trails modeling, where user trails record sequences of activities when individuals interact with the Internet and the world. We propose the retrospective higher-order Markov process as a two-step process by first choosing a state from the history and then transitioning as a first-order chain conditional on that state. This way the total number of parameters is restricted and thus the model is protected from overfitting. Lastly we propose to use a time-inhomogeneous Markov chain to approximate the solution of a linear system. Multiple simulations of the random walk are conducted to approximate the solution. By allowing the random walk to transition based on multiple matrices, we decrease the variance of the simulations, and thus increase the speed of the solver.</p><p>