Eva Bozich Brundage

Future implementation of this GUI would be allowed to interact with other system in the household with the intent to eliminate the pollutants for the home owners without any action from the owner.

The goal of this project is to use Indoor Air Data to create an interactive visualization representative of the indoor air quality. The visualizations built focus on representing the current levels of pollutants in the air. There is also promise in using machine learning to predict the toxicity of indoor air to help users optimize the air quality.

One of the founding fathers of Machine Learning, Yann LeCun, said, "There are many interesting recent developments in deep learning… The most important one, in my opinion, is adversarial training, also called GAN for Generative Adversarial Networks. This, and the variations that are now being proposed, is the most interesting idea in the last 10 years in Machine Leaning." Our CycleGAN model uses two GANs simultaneously, learning to generate photos with unpaired image datasets.

Visit Medium for a full description. The past five years of computer vision have been expanded by a large factor through the use of a technique called Deep Learning. Deep Learning, concerns with the utility of artificial neural networks in use. One big leap in computer vision and deep learning is the concept of Generative Adversarial Networks, which consists of two competing models, one of them trying to learn a correct image through noise, meanwhile the second model evaluating weather the other did a good job or still has time to go. Paired image datasets are time consuming to curate, they are expensive to curate, and it is likely impossible to find the exact dataset required to carry out of the desired implementation. CycleGANs solves that problem of paired data opening a brand new branch in Machine Learning: unsupervised image to image translation.

This research digs into whether Computer Engineers should be held as responsible for their work as engineers in others disciplines.

While most engineering disciplines have strict codes of conduct that must be followed to the tee, the Computer Science/Engineering discipline is relatively new and laws are not established when it comes to regulating coding practices.

Research was done using a “four pillar approach” with the following categories: Code is Law; Software as a Sociotechnical System; Questioning the Engineering Mindset; Claiming Ethical Neutrality.

An Air Motor complete with CREO modeling and technical drawings and manufacturing specifications.

Hand drafted the specifications for this Air Motor. Modeled the specifications of the Air Motor in CREO Parametric. Crafted the components of this Air Motor using a mill, lathe, and CNC machine.

Constructed and Flew an RC Glider.

Built using blue foam, balsa wood, and epoxy. Pennies were on board… and the Glider handled their load well… For a few seconds.

Constructed and Launched a Rocket

Built using under $10 worth of materials including plastic Smart Water bottle, and a plastic paint scraper. Built and launched a rocket with an egg inside… it crashed… but the egg survived and no it was not hard boiled.