Digital Commons Network^™

Predicting Changes In Earnings: A Walk Through A Random Forest, Joshua Hunt

Graduate Theses and Dissertations

This paper investigates whether the accuracy of models used in accounting research to predict categorical dependent variables (classification) can be improved by using a data analytics approach. This topic is important because accounting research makes extensive use of classification in many different research streams that are likely to benefit from improved accuracy. Specifically, this paper investigates whether the out-of-sample accuracy of models used to predict future changes in earnings can be improved by considering whether the assumptions of the models are likely to be violated and whether alternative techniques have strengths that are likely to make them a better choice …

Improving Asynchronous Advantage Actor Critic With A More Intelligent Exploration Strategy, James B. Holliday

Graduate Theses and Dissertations

We propose a simple and efficient modification to the Asynchronous Advantage Actor Critic (A3C)

algorithm that improves training. In 2016 Google’s DeepMind set a new standard for state-of-theart

reinforcement learning performance with the introduction of the A3C algorithm. The goal of

this research is to show that A3C can be improved by the use of a new novel exploration strategy we

call “Follow then Forage Exploration” (FFE). FFE forces the agents to follow the best known path

at the beginning of a training episode and then later in the episode the agent is forced to “forage”

and explores randomly. In …

A Home Security System Based On Smartphone Sensors, Michael Mahler

Graduate Theses and Dissertations

Several new smartphones are released every year. Many people upgrade to new phones, and their old phones are not put to any further use. In this paper, we explore the feasibility of using such retired smartphones and their on-board sensors to build a home security system. We observe that door-related events such as opening and closing have unique vibration signatures when compared to many types of environmental vibrational noise. These events can be captured by the accelerometer of a smartphone when the phone is mounted on a wall near a door. The rotation of a door can also be captured …

Improving The Efficacy Of Context-Aware Applications, Jon C. Hammer

Graduate Theses and Dissertations

In this dissertation, we explore methods for enhancing the context-awareness capabilities of modern computers, including mobile devices, tablets, wearables, and traditional computers. Advancements include proposed methods for fusing information from multiple logical sensors, localizing nearby objects using depth sensors, and building models to better understand the content of 2D images.

First, we propose a system called Unagi, designed to incorporate multiple logical sensors into a single framework that allows context-aware application developers to easily test new ideas and create novel experiences. Unagi is responsible for collecting data, extracting features, and building personalized models for each individual user. We demonstrate the …

Baseline Data From Servo Motors In A Robotic Arm For Autonomous Machine Fault Diagnosis, Jacob Brown

Mechanical Engineering Undergraduate Honors Theses

Fault diagnosis can prolong the life of machines if potential sources of failure are discovered and corrected before they occur. Supervised machine learning, or the use of training data to enable machines to discover these faults on their own, makes failure prevention much easier. The focus of this thesis is to investigate the feasibility of creating datasets of various faults at both the component and system level for a servomotor and a compatible robotic arm, such that this data can be used in machine learning algorithms for fault diagnosis. The faults induced at the component level in different servomotors include: …

A Continuous Space Generative Model, Erzen Komoni

Graduate Theses and Dissertations

Generative models are a class of machine learning models capable of producing digital images with plausibly realistic properties. They are useful in such applications as visualizing designs, rendering game scenes, and improving images at higher magnifications. Unfortunately, existing generative models generate only images with a discrete predetermined resolution. This paper presents the Continuous Space Generative Model (CSGM), a novel generative model capable of generating images as a continuous function, rather than as a discrete set of pixel values. Like generative adversarial networks, CSGM trains by alternating between generative and discriminative steps. But unlike generative adversarial networks, CSGM uses only one …

Training Machine Learning Agents In A 3d Game Engine, Diego Calderon

Computer Science and Computer Engineering Undergraduate Honors Theses

Artificial intelligence (AI) and video games benefit from each other. Games provide a challenging domain for testing learning algorithms, and AI provides a framework to designing and implementing intelligent behavior, which reinforces meaningful play. Medium and small studios, and independent game developers, have limited resources to design, implement, and maintain agents with reactive behavior. In this research, we trained agents using machine learning (ML), aiming to find an alternative to expensive traditional algorithms for intelligent behavior used in video games. We use Unity as a game engine to implement the environments and TensorFlow for the neural network training.