Filtering by
- All Subjects: Machine Learning
- Creators: Barrett, The Honors College
The aim of this creative project was to explore the ideas of impermanence and transience through the lens of different, largely non-western cultural backgrounds, and to incorporate what I learned into my own work as a painter. As part of this, I focused on the materials, techniques, visual strategies, and philosophies that guided the creation of these works. The project consisted of a discrete research phase, during which time I gathered information and materials related to my topic, and a creation phase, when I focused largely on the production of oil paintings and ink paintings whose technique and/or subject matter pertained to impermanence. Research for the most part was conducted by utilizing online and physical collections of work to analyze the formal elements of the work along with the cultural context in which it was created. Ultimately the creative project resulted in a product of three oil paintings and five ink paintings.
The research presented in this Honors Thesis provides development in machine learning models which predict future states of a system with unknown dynamics, based on observations of the system. Two case studies are presented for (1) a non-conservative pendulum and (2) a differential game dictating a two-car uncontrolled intersection scenario. In the paper we investigate how learning architectures can be manipulated for problem specific geometry. The result of this research provides that these problem specific models are valuable for accurate learning and predicting the dynamics of physics systems.<br/><br/>In order to properly model the physics of a real pendulum, modifications were made to a prior architecture which was sufficient in modeling an ideal pendulum. The necessary modifications to the previous network [13] were problem specific and not transferrable to all other non-conservative physics scenarios. The modified architecture successfully models real pendulum dynamics. This case study provides a basis for future research in augmenting the symplectic gradient of a Hamiltonian energy function to provide a generalized, non-conservative physics model.<br/><br/>A problem specific architecture was also utilized to create an accurate model for the two-car intersection case. The Costate Network proved to be an improvement from the previously used Value Network [17]. Note that this comparison is applied lightly due to slight implementation differences. The development of the Costate Network provides a basis for using characteristics to decompose functions and create a simplified learning problem.<br/><br/>This paper is successful in creating new opportunities to develop physics models, in which the sample cases should be used as a guide for modeling other real and pseudo physics. Although the focused models in this paper are not generalizable, it is important to note that these cases provide direction for future research.
High-entropy alloys possessing mechanical, chemical, and electrical properties that far exceed those of conventional alloys have the potential to make a significant impact on many areas of engineering. Identifying element combinations and configurations to form these alloys, however, is a difficult, time-consuming, computationally intensive task. Machine learning has revolutionized many different fields due to its ability to generalize well to different problems and produce computationally efficient, accurate predictions regarding the system of interest. In this thesis, we demonstrate the effectiveness of machine learning models applied to toy cases representative of simplified physics that are relevant to high-entropy alloy simulation. We show these models are effective at learning nonlinear dynamics for single and multi-particle cases and that more work is needed to accurately represent complex cases in which the system dynamics are chaotic. This thesis serves as a demonstration of the potential benefits of machine learning applied to high-entropy alloy simulations to generate fast, accurate predictions of nonlinear dynamics.
Robots are often used in long-duration scenarios, such as on the surface of Mars,where they may need to adapt to environmental changes. Typically, robots have been built specifically for single tasks, such as moving boxes in a warehouse or surveying construction sites. However, there is a modern trend away from human hand-engineering and toward robot learning. To this end, the ideal robot is not engineered,but automatically designed for a specific task. This thesis focuses on robots which learn path-planning algorithms for specific environments. Learning is accomplished via genetic programming. Path-planners are represented as Python code, which is optimized via Pareto evolution. These planners are encouraged to explore curiously and efficiently. This research asks the questions: “How can robots exhibit life-long learning where they adapt to changing environments in a robust way?”, and “How can robots learn to be curious?”.
Colorimetric assays are an important tool in point-of-care testing that offers several advantages to traditional testing methods such as rapid response times and inexpensive costs. A factor that currently limits the portability and accessibility of these assays are methods that can objectively determine the results of these assays. Current solutions consist of creating a test reader that standardizes the conditions the strip is under before being measured in some way. However, this increases the cost and decreases the portability of these assays. The focus of this study is to create a machine learning algorithm that can objectively determine results of colorimetric assays under varying conditions. To ensure the flexibility of a model to several types of colorimetric assays, three models were trained on the same convolutional neural network with different datasets. The images these models are trained on consist of positive and negative images of ETG, fentanyl, and HPV Antibodies test strips taken under different lighting and background conditions. A fourth model is trained on an image set composed of all three strip types. The results from these models show it is able to predict positive and negative results to a high level of accuracy.
The field of biomedical research relies on the knowledge of binding interactions between various proteins of interest to create novel molecular targets for therapeutic purposes. While many of these interactions remain a mystery, knowledge of these properties and interactions could have significant medical applications in terms of understanding cell signaling and immunological defenses. Furthermore, there is evidence that machine learning and peptide microarrays can be used to make reliable predictions of where proteins could interact with each other without the definitive knowledge of the interactions. In this case, a neural network was used to predict the unknown binding interactions of TNFR2 onto LT-ɑ and TRAF2, and PD-L1 onto CD80, based off of the binding data from a sampling of protein-peptide interactions on a microarray. The accuracy and reliability of these predictions would rely on future research to confirm the interactions of these proteins, but the knowledge from these methods and predictions could have a future impact with regards to rational and structure-based drug design.
The Student Art Project is art patronage for the 21st century—a curated online gallery featuring exceptional student artists. The Student Art Project is a highly curated experience for buyers. Only five artists are featured each month. Buyers are not bombarded with thousands of different products and separate artists “shops”. They can read artists bios and find art they connect with.
Student artists apply through an online form. Once accepted to the program, artists receive a $200 materials stipend to create an exclusive collection of 5-10 pieces. Original artwork and limited edition prints are sold through our website. These collections can potentially fund an entire year of college tuition, a life-changing amount for many students.
Brick-and-mortar galleries typically take 40-60% of the retail price of artwork. The Student Art Project will only take 30%, which we will use to reinvest in future artists. Other art websites, like Etsy, require the artists to ship, invoice, and communicate with customers. For students, this means less time spent in the classroom and less time developing their craft. The Student Art Project handles all business functions for our artists, allowing them to concentrate on what really matters, their education.
Anthropocentric society faces a multiplicity of environmental challenges, catalyzed and perpetuated by urban-industrial culture. Many of today’s perspectives and sustainable strategies cannot accommodate the challenges’ inherent complexity. Because urban-industrial society is only projected to grow, both in enormity and influence, the only viable option is to elucidate the complexity and employ it.
A potential setting in which to frame this exploration is the intersection of urbanism, landscape, and ecology –an overlap first introduced by the theories of Landscape Urbanism and Ecological Urbanism. Here, urbanization is not just discussed as an isolated phenomenon but one that is embedded within and responding to a variety of systems and scales. The methodologies of Landscape Urbanism and Ecological Urbanism also acknowledge artists and the visual arts as invaluable tools for realizing, communicating, and inspiring the new perspectives and modes of intervention needed to address the aforementioned urban complexity. Such artists who operate within this realm include Sissel Tolaas, Maya Lin, Katrin Sigurdardottir, David Maisel, Olafur Eliason, Mierle Ukeles, Suzanne Lacy, Steve Rowell, Mel Chin, and the Center for Land Use Interpretation. Case study analyses reveal many of these artists begin their investigations with provocative, searching questions situated within the realms of urbanism, landscape, and ecology. This is proceeded by relative scientific research and/or community involvement or outreach. Furthermore, the artists work within and extrapolate from a variety of other disciplines —increasing the scope and applicability of their work. The information they collect via this multidisciplinary approach is then metaphorically translated to the visual arts, where the public can not only physically or sensorially experience it, but understand and deduce its meaning and significance: public awareness being one of the more essential aspects of a sustainable society and at the root of our current struggle.
As a designer and architect, I will engage the artist’s mindset to explore the current and complex issue of resource extraction within Superior, Arizona: a topic at the core of urbanism, landscape, and ecology. While the town is not considered "urban" by standard definition, it and its surrounding landscapes are indirectly sculpted by the needs of urban society —rendering it the setting for this application. Within a group, we will begin with a searching question. We will conduct relative scientific research, engage the community of Superior, and call upon a variety of other disciplines to aid and inform our work. Through metaphor, the research and resulting discoveries will be artistically represented and composed within a designed exhibition of hopeful “things” (See Bruno Latour, “From Realpolitik to Dingpolitik”). This exhibition will theoretically take place on Superior’s currently dilapidated Main Street, amid a more accessible sphere. The eventual goal of the project is to illuminate and understand the complexities of resource extraction, specifically within Superior, while also enabling public awareness and empowerment through lucidity and comprehension.