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.

The depiction of female action heroes in modern blockbuster films has become more accurate throughout the years; however, the representation of women is still not as progressive as feminist scholars, actors, and viewers would like. This thesis explores two recent blockbuster films, Wonder Woman (2017) and Mad Max: Fury Road (2015), and how each film deals with the representation of women. While one could look to many cultural forms to explore such issues, films, “the most accessible representations of the past, present, and future of our society,” are particularly fertile ground for exploring gendered representations and stereotyping (Haskell, 1974). For much of Hollywood history, action films have used female protagonists as either passive, venerated symbols of perfect femininity, or objects of fascination and sexual pleasure for their male viewers. Or, if the female hero does have a degree of agency that allows her to push the plot forward, she is subject to moral scrutiny and frequently masculinized. In fact, the representation of women often falls into binary categories: the angelic damsel in distress, or the morally reprehensible, often masculinized, female villain. While the history of women’s representation in film more generally and action films more specifically is a long and complicated one that is beyond the scope of this project, recent action productions have exhibited notable shifts, both in terms of female characters’ box-office and narrative strength. However, both Wonder Woman and Mad Max: Fury Road, present viewers with examples of female representation that break through many of the misogynistic tropes that have dominated the genre for far too long. The key distinction between how both films destroy gendered stereotypes lies in the degrees to which the films allow their central female protagonists, and more minor female characters, to dominate the narrative and inhabit the composition of the screen. Wonder Woman tells the story of one powerful woman, whereas Fury Road utilizes a multitude of women in its story to defy gender stereotypes. While both films can be interpreted as empowering for female viewers, Wonder Woman gives its audience an easily digestible example of female agency; this is due to Wonder Woman allowing its famous comic book hero to comment and reject traditional women’s clothing, but also insists Diana be limited to hypersexualized battle armour and implicates that women cannot have love, power, and family. On the other hand, Fury Road presents viewers with a more radicalized gynocentric world in which, after considerable struggle and not without compromise, female characters not only have power, but wrest it away from the men who have abusively held onto control in the past. These two films also paved new ground for women in Hollywood production terms: giving women more power at the box-office and destroying the old-aged notion that female-centric films do not sell and make money at the same rate as male-centered ones do. Both Wonder Woman and Mad Max: Fury Road, in their own ways, depict that there is space for female action heroes to be more progressive and feminist in future blockbuster action films.

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.

Filmmakers seek to create story pieces that are visually beautiful and engage the full attention of their audience. They typically abide by a 3-step process moving through pre-production, production, and post-production. Within each step, there are a series of tasks that need to be accomplished in order to reach the completed film. A successful film requires careful planning and strategy in pre-production, timely and decisive execution in production, and minimal unforeseen retouching in post-production.<br/><br/>Even though filmmakers have continued to follow the same formula throughout the decades, the filmmaking process has remained largely inefficient. It is extremely common for pre-production planning to be undercut, for production filming to run far too long, and for post-production VFX and editing to send the project over budget. These instances can cause major issues as the project is being finalized. In many scenarios portions of the project need to be reshot, the box office revenue isn’t enough to make up for extensive VFX retouching, or the project may never even come to fruition. <br/><br/>The reason for this recurring theme of films being over budget and out of time is quite simply that technology has made filmmakers lazy. “Fix it in post” is a disgustingly common phrase used in the film industry. It describes the utter abuse of computer retouching in the post-production phase of filmmaking. Despite working in an industry that seeks to entertain the human eye, filmmakers have become blind to all of the small mistakes that could cost them hundreds of hours and millions of dollars in the long run.

Through this creative project, I analyzed how COVID-19 has affected the theatre industry. I created a mini-documentary following ASU’s production of Runaways, which was performed without an audience. The final product was a combination of pre-filmed and self-taped scenes. I documented how students were still able to learn and cultivate their skills during a time where most things are virtual. In addition, I analyzed how the shift to filmed theatre has changed the definition of live theatre, including increased accessibility. I also explored the importance of theatre through analyzing the themes of musical theatre performances such as Rent and Runaways. During a time where people cannot gather, artists are still finding a way to create and tell stories.

Can You Hear Me is a short documentary which seeks to give voice to the experiences of trans and nonbinary students in ASU classrooms. What I present in this project are the direct spoken accounts of the feelings, thoughts and frustrations of transgender and nonbinary students as they navigate university classrooms at Arizona State University. Can You Hear Me serves as a representational platform for trans and nonbinary students to communicate their experiences to other students, staff and faculty in the hopes that it might help make classroom spaces more inclusive.

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.