Deforestation in the Amazon rainforest has the potential to have devastating effects on ecosystems on both a local and global scale, making it one of the most environmentally threatening phenomena occurring today. In order to minimize deforestation in the Ama- zon and its consequences, it is helpful to analyze its occurrence using machine learning architectures such as the U-Net. The U-Net is a type of Fully Convolutional Network that has shown significant capability in performing semantic segmentation. It is built upon a symmetric series of downsampling and upsampling layers that propagate feature infor- mation into higher spatial resolutions, allowing for the precise identification of features on the pixel scale. Such an architecture is well-suited for identifying features in satellite imagery. In this thesis, we construct and train a U-Net to identify deforested areas in satellite imagery of the Amazon through semantic segmentation.

Our Idea: As a team of engineers, two in the engineering field and one in computer science and software development, we wanted to find a way to put these skills to use in our company. As we did not have a revolutionary idea to build our own product, we wanted to base our company on the assumption that people have great ideas and lack the ability to execute on these ideas. Our mission is to enable these people and companies to make their ideas a reality, and allow them to go to market with a clean and user friendly product. We are using our skills and experience in hardware and device prototyping and testing, as well as software design and development to make this happen. Implementation: To this point, we have been working with a client building a human diagnostic and enhancement AI device. We have been consulting on mostly the design and creation of their first proof of concept, working on hardware and sensor interaction as well as developing the software allowing their platform to come to life. We have been working closely with the leaders of the company, who have the ideas and business knowledge, while we focus on the technology side. As for the scalability and market potential of our business, we believe that the potential market is not the limiting factor. Instead, the limiting factor to the growth of our business is the time we have to devote. We are currently only working with one client, and not looking to expand into new clients. We believe this would require the addition of new team members, but instead we are happy with the progress we are making at the moment. We believe we are not only building equity in business we believe in, but also building a product that could help the safety and wellness of our users.

In the field of botany, it is often necessary for plants to be identified based on their phenotypical characteristics, whether in person or using previously collected image samples. This work can be tedious and challenging for a human botanist to complete, as datasets can be large and several species of plants strongly resemble each other. Various machine learning techniques, both supervised and unsupervised, can address this task with varying degrees of accuracy and efficiency thanks to their ability to identify subtle patterns in data. The objective of this research is to both conduct a review of previous studies that measure the effectiveness of various machine learning methods for plant identification and to build and test various models to draw up a comparison of the accuracies and efficiencies of the set of techniques. A review of the existing literature found that any of the studied machine learning techniques can yield a high level of accuracy when used in the correct situations and on a suitable dataset. The results gathered from the models built from this research show that all else being equal, complex convolutional neural networks perform the best on this task, yielding an accuracy of 85.4% on the larger dataset. The other models tested in descending order of accuracy on the same dataset are k-nearest neighbors, random forest, k-means clustering, and a decision tree classifier.

This review explores popular gambling strategies often believed to guarantee wins, such as card counting and taking advantage of arbitrage. We present a mathematical overview of these systems to evaluate their theoretical effectiveness in ideal conditions by presenting prior research and mathematical proofs. This paper then generates results from these models using Monte Carlo simulations and compares them to data from real-world scenarios. Additionally, we examine reasons that might explain the discrepancies between theoretical and real-world results, such as the potential for dealers to detect and counteract card counting. Ultimately, although these strategies may fare well in theoretical scenarios, they struggle to create long-term winning solutions in casino or online gambling settings.