Physical Sciences and Mathematics Commons^™

Identifying Key Activity Indicators In Rats' Neuronal Data Using Lasso Regularized Logistic Regression, Avery Woods

Honors Theses

This thesis aims to identify timestamps of rats’ neuronal activity that best determine behavior using a machine learning model. Neuronal data is a complex and high-dimensional dataset, and identifying the most informative features is crucial for understanding the underlying neuronal processes. The Lasso regularization technique is employed to select the most relevant features of the data to the model’s prediction. The results of this study provide insights into the key activity indicators that are associated with specific behaviors or cognitive processes in rats, as well as the effect that stress can have on neuronal activity and behavior. Ultimately, it was …

Digital Dna: The Ethical Implications Of Big Data As The World’S New-Age Commodity, Clark H. Dotson

Honors Theses

In the emerging digital world that we find ourselves in, it becomes apparent that data collection has become a staple of daily life, whether we like it or not. This research discussion aims to bring light to just how much one’s own digital identity is valued in the technologically-infused world of today, with distinct research and local examples to bring awareness to the ethical implications of your online presence. The paper in question examines anecdotal and research evidence of the collection of data, both through true and unjust means, as well as ethical implications of what this information truly represents. …

Impact Of Movements On Facial Expression Recognition, Zhebin Yin

Honors Theses

The ability to recognize human emotions can be a useful skill for robots. Emotion recognition can help robots understand our responses to robot movements and actions. Human emotions can be recognized through facial expressions. Facial Expression Recognition (FER) is a well-established research area, how- ever, the majority of prior research is based on static datasets of images. With robots often the subject is moving, the robot is moving, or both. The purpose of this research is to determine the impact of movement on facial expression recognition. We apply a pre-existing model for FER, which performs around 70.86% on a given …

The Future Of Artificial Intelligence In The Healthcare Industry, Erika Bonnist

Honors Theses

Technology has played an immense role in the evolution of healthcare delivery for the United States and on an international scale. Today, perhaps no innovation offers more potential than artificial intelligence. Utilizing machine intelligence as opposed to human intelligence for the purposes of planning, offering solutions, and providing insights, AI has the ability to alter traditional dynamics between doctors, patients, and administrators; this reality is now producing both elation at artificial intelligence's medical promise and uncertainty regarding its capacity in current systems. Nevertheless, current trends reveal that interest in AI among healthcare stakeholders is continuously increasing, and with the current …

Using Deep Learning To Automate The Diagnosis Of Skin Melanoma, Akhil Reddy Alasandagutti

Honors Theses

Machine learning and image processing techniques have been widely implemented in the field of medicine to help accurately diagnose a multitude of medical conditions. The automated diagnosis of skin melanoma is one such instance. However, a majority of the successful machine learning models that have been implemented in the past have used deep learning approaches where only raw image data has been utilized to train machine learning models, such as neural networks. While they have been quite effective at predicting the condition of these lesions, they lack key information about the images, such as clinical data, and features that medical …

Variable Autoencoders For Biosensor Data Augmentation, Solomon Kim

Honors Theses

Over the past decade machine learning and artificial intelligence's resurgence spawned the desire to mimic human creative ability. Initially attempts to create images, music, and text flooded the community, though little has been learned regarding constrained, one-dimensional data generation. This paper demonstrates a variational autoencoder approach to this problem. By modeling biosensor current and concentration data we aim to augment the existing dataset. In training a multi-layer neural network based encoder and decoder we were able to generate realistic, original samples., These results demonstrate the ability to realistically augment datasets, improving training of machine learning models designed to predict concentration …

Estimating Heading From Optic Flow With Neural Networks, Natalie T. Maus

Honors Theses

Humans have a remarkable ability to estimate their direction of self-motion, or heading, based on visual input stimulus (optic flow). Machines, on the other hand, have a difficult time with this task, especially when flow is introduced that is inconsistent with the motion of the observer. For example, when moving objects enter the field of view, their motion provides inconsistent flow data which often disrupts heading estimates of current heading estimation models. We investigate the ability of neural networks to estimate heading from optic flow data and the limitations of these models when different variations of inconsistent flow are introduced. …

Convolutional Audio Source Separation Applied To Drum Signal Separation, Marius Orehovschi

Honors Theses

This study examined the task of drum signal separation from full music mixes via both classical methods (Independent Component Analysis) and a combination of Time-Frequency Binary Masking and Convolutional Neural Networks. The results indicate that classical methods relying on predefined computations do not achieve any meaningful results, while convolutional neural networks can achieve imperfect but musically useful results. Furthermore, neural network performance can be improved by data augmentation via transposition – a technique that can only be applied in the context of drum signal separation.

Perceptually Improved Medical Image Translations Using Conditional Generative Adversarial Networks, Anurag Vaidya

Honors Theses

Magnetic resonance imaging (MRI) can help visualize various brain regions. Typical MRI sequences consist of T1-weighted sequence (favorable for observing large brain structures), T2-weighted sequence (useful for pathology), and T2-FLAIR scan (useful for pathology with suppression of signal from water). While these different scans provide complementary information, acquiring them leads to acquisition times of ~1 hour and an average cost of $2,600, presenting significant barriers. To reduce these costs associated with brain MRIs, we present pTransGAN, a generative adversarial network capable of translating both healthy and unhealthy T1 scans into T2 scans. We show that the addition of non-adversarial …

Applying Deep Learning For Cell Detection In Time-Lapse Microscopic Images, Jay Patel

Honors Theses

The budding yeast Saccharomyces cerevisiae is an effective model for studying cellular aging. We can measure the lifespan of yeast cells in two ways: replicative and chronological lifespans. Chronological focuses on the time that a cell can survive. The replicative lifespan (RLS) is the number of cell divisions that a single mother cell can go through before ceases to be dividing. RLS is a measurement of individual cells and is more informative on the aging process than in chronological lifespan. Many genes that influence yeast RLS have been shown to be highly conserved and have a similar effect on aging …

Human Facial Emotion Recognition System In A Real-Time, Mobile Setting, Claire Williamson

Honors Theses

The purpose of this project was to implement a human facial emotion recognition system in a real-time, mobile setting. There are many aspects of daily life that can be improved with a system like this, like security, technology and safety.

There were three main design requirements for this project. The first was to get an accuracy rate of 70%, which must remain consistent for people with various distinguishing facial features. The second goal was to have one execution of the system take no longer than half of a second to keep it as close to real time as possible. Lastly, …

A Machine Learning Method For Predicting Liver Transplant Survival Outcomes, Brandon C. Revels

Honors Theses

For years, doctors have utilized the Model for End-stage Liver Disease (MELD) score to aid in the allocation of organs for liver transplants (LT). A major issue with using the MELD score to allocate organs for transplantation is that the MELD score does not accurately predict post-transplant survival. This research project aims to investigate the use of machine learning (ML) methods to predict LT survival using the newer Scientific Registry of Transplant Recipients (SRTR) dataset. For this project, death and nonfatal graft failure were treated equally as both cases result in a loss of a donated organ. The ML algorithms …

Exploring The Behavior Repertoire Of A Wireless Vibrationally Actuated Tensegrity Robot, Zongliang Ji

Honors Theses

Soft robotics is an emerging field of research due to its potential to explore and operate in unstructured, rugged, and dynamic environments. However, the properties that make soft robots compelling also make them difficult to robustly control. Here at Union, we developed the world’s first wireless soft tensegrity robot. The goal of my thesis is to explore effective and efficient methods to explore the diverse behavior our tensegrity robot. We will achieve that by applying state-of-art machine learning technique and a novelty search algorithm.

Designing A General Education Course On The Societal Impacts Of Artificial Intelligence, Vincent Rollins

Honors Theses

Most colleges, including UTC, already offer an artificial intelligence course (CPSC 4440) as part of their computer science curricula. Such courses are meant to explain the technology behind these elaborate systems, but these courses often neglect extensive coverage of the real-world impacts of the technology itself. UTC also offers a course entitled “Ethical and Social Issues in Computing” that does convey the importance behind the advances of computer technology and its impacts, but this course is practically available only to computer science majors. There is no generalized and widely available course that covers the technological, economic, cultural, philosophical/theological, and ethical …

Bridging Act-R And Project Malmo, Developing Models Of Behavior In Complex Environments, David M. Schwartz

Honors Theses

Cognitive architectures such as ACT-R provide a system for simulating the mind and human behavior. On their own they model decision making of an isolated agent. However, applying a cognitive architecture to a complex environment yields more interesting results about how people make decisions in more realistic scenarios. Furthermore, cognitive architectures enable researchers to study human behavior in dangerous tasks which cannot be tested because they would harm participants. Nonetheless, these architectures aren’t commonly applied to such environments as they don’t come with one. It is left to the researcher to develop a task environment for their model. The difficulty …

2018 Ieee Signal Processing Cup: Forensic Camera Model Identification Challenge, Michael Geiger

Honors Theses

The goal of this Senior Capstone Project was to lead Union College’s first ever Signal Processing Cup Team to compete in IEEE’s 2018 Signal Processing Cup Competition. This year’s competition was a forensic camera model identification challenge and was divided into two separate stages of competition: Open Competition and Final Competition. Participation in the Open Competition was open to any teams of undergraduate students, but the Final Competition was only open to the three finalists from Open Competition and is scheduled to be held at ICASSP 2018 in Calgary, Alberta, Canada. Teams that make it to the Final Competition will …

Optimizing Tensegrity Gaits Using Bayesian Optimization, James Boggs

Honors Theses

We design and implement a new, modular, more complex tensegrity robot featuring data collection and wireless communication and operation as well as necessary accompanying research infrastructure. We then utilize this new tensegrity to assess previous research on using Bayesian optimization to generate effective forward gaits for tensegrity robots. Ultimately, we affirm the conclusions of previous researchers, demonstrating that Bayesian optimization is statistically significantly (p < 0:05) more effective at discovering useful gaits than random search. We also identify several flaws in our new system and identify means of addressing them, paving the way for more effective future research.

Self-Reconfiguration Planning In Modular Reconfigurable Robots, Keaton Griffith

Honors Theses

MSRs are highly versatile robots that work together to form into different configurations. However, to take advantage of this ability to transform, the MSR must utilize an SRP algorithm to determine what actions to perform to shape itself to reach its goal configuration. An SRP algorithm can be boiled down to a search method through an unexplored graph which we approach with four basic search algorithms to see which algorithm is best when designing an SRP algorithm. To do this we create a general MSR model known as stickbots and use different search algorithms on a variety of SRP problems …

A Better Way To Construct Tensegrities: Planar Embeddings Inform Tensegrity Assembly, Elizabeth Anne Ricci

Honors Theses

Although seemingly simple, tensegrity structures are complex in nature which makes them both ideal for use in robotics and difficult to construct. We work to develop a protocol for constructing tensegrities more easily. We consider attaching a tensegrity's springs to the appropriate locations on some planar arrangement of attached struts. Once all of the elements of the structure are connected, we release the struts and allow the tensegrity to find its equilibrium position. This will allow for more rapid tensegrity construction. We develop a black-box that given some tensegrity returns a flat-pack, or the information needed to perform this physical …

Attracting Human Attention Using Robotic Facial Expressions And Gestures, Venus Yu

Honors Theses

Robots will soon interact with humans in settings outside of a lab. Since it will be likely that their bodies will not be as developed as their programming, they will not have the complex limbs needed to perform simple tasks. Thus they will need to seek human assistance by asking them for help appropriately. But how will these robots know how to act? This research will focus on the specific nonverbal behaviors a robot could use to attract someone’s attention and convince them to interact with the robot. In particular, it will need the correct facial expressions and gestures to …

Effective Ann Topologies For Use As Genotypes For Evaluating Design And Fabrication, John R. Peterson

Honors Theses

There is promise in the field of Evolutionary Design for systems that evolve not only what to manufacture but also how to manufacture it. EvoFab is a system that uses Genetic Algorithms to evolve Artificial Neural Networks (ANNs) which control a modified 3d-printer with the goal of automating some level of invention. ANNs are an obvious choice for use with a system like this as they are canonically evolvable encodings, and have been successfully used as evolved control systems in Evolutionary Robotics. However, there is little known about how the structural characteristics of an ANN affect the shapes that can …

An Alternative Approach To Training Sequence-To-Sequence Model For Machine Translation, Vivek Sah

Honors Theses

Machine translation is a widely researched topic in the field of Natural Language Processing and most recently, neural network models have been shown to be very effective at this task. The model, called sequence-to-sequence model, learns to map an input sequence in one language to a vector of fixed dimensionality and then map that vector to an output sequence in another language without any human intervention provided that there is enough training data. Focusing on English-French translation, in this paper, I present a way to simplify the learning process by replacing English input sentences by word-by-word translation of those sentences. …

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 …

Using Genetic Algorithms To Evolve Artificial Neural Networks, William T. Kearney

Honors Theses

This paper demonstrates that neuroevolution is an effective method to determine an optimal neural network topology. I provide an overview of the NeuroEvolution of Augmenting Topologies (NEAT) algorithm, and describe how unique characteristics of this algorithm solve various problem inherent to neuroevolution (namely the competing conventions problem and the challenges associated with protecting topological innovation). Parallelization is shown to greatly speed up efficiency, further reinforcing neuroevolution as a potential alternative to traditional backpropagation. I also demonstrate that appropriate parameter selection is critical in order to efficiently converge to an optimal topology. Lastly, I produce an example solution to a medical …

What It Is To Be Conscious: Exploring The Plasibility Of Consciousness In Deep Learning Computers, Peter Davis

Honors Theses

As artificial intelligence and robotics progress further and faster every day, designing and building a conscious computer appears to be on the horizon. Recent technological advances have allowed engineers and computer scientists to create robots and computer programs that were previously impossible. The development of these highly sophisticated robots and AI programs has thus prompted the age-old question: can a computer be conscious? The answer relies on addressing two key sub-problems. The first is the nature of consciousness: what constitutes a system as conscious, or what properties does consciousness have? Secondly, does the physical make-up of the robot or computer …

The Advanced Educational Robot, Calder Phillips-Grafflin

Honors Theses

Existing literature in the field of computer science education clearly demonstrates that robots can be ideal teaching tools for basic computer science concepts. Likewise, robots are an ideal platform for more complicated CS techniques such as evolutionary algorithms and neural networks. With these two distinct roles in mind, that of the teaching tool and that of the research tool, in collaboration with customers in the CS department we have developed a new robotics platform suitable for both roles that provides higher performance and improved ease-of-use in comparison to the robots currently in use at Union. We have successfully designed and …

Masterpiece: Computer-Generated Music Through Fractals And Genetic Theory, Amanda Broyles

Honors Theses

A wide variety of computer-generated music exists. I have writ.ten a program which will generate music by using genetic theory and fractals. The genetic theory is used to mold input pieces into a musical motif. The motif is then elaborated by the fractal formula into a composition. A brief introduction to the world of genetic theory and fractals is given. Analysis of a musical work produced in this manner shows coherent patterns and also emotion.