Matching Items (1,137)
Filtering by
- Creators: School of Mathematical and Statistical Sciences
- Creators: Department of Finance
- Member of: Theses and Dissertations
Description
The performance of modern machine learning algorithms depends upon the selection
of a set of hyperparameters. Common examples of hyperparameters are learning
rate and the number of layers in a dense neural network. Auto-ML is a branch
of optimization that has produced important contributions in this area. Within
Auto-ML, multi-fidelity approaches, which eliminate poorly-performing configurations
after evaluating them at low budgets, are among the most effective. However, the
performance of these algorithms strongly depends on how effectively they allocate
the computational budget to various hyperparameter configurations. We first present
Parameter Optimization with Conscious Allocation 1.0 (POCA 1.0), a hyperband-
based algorithm for hyperparameter optimization that adaptively allocates the inputted
budget to the hyperparameter configurations it generates following a Bayesian sampling
scheme. We then present its successor Parameter Optimization with Conscious
Allocation 2.0 (POCA 2.0), which follows POCA 1.0’s successful philosophy while
utilizing a time-series model to reduce wasted computational cost and providing a
more flexible framework. We compare POCA 1.0 and 2.0 to its nearest competitor BOHB
at optimizing the hyperparameters of a multi-layered perceptron and find that both
POCA algorithms exceed BOHB in low-budget hyperparameter optimization while
performing similarly in high-budget scenarios.
ContributorsInman, Joshua (Author) / Sankar, Lalitha (Thesis director) / Pedrielli, Giulia (Committee member) / Barrett, The Honors College (Contributor) / School of Mathematical and Statistical Sciences (Contributor) / Computer Science and Engineering Program (Contributor)
Created2024-05
Description
With the growing presence of Artificial Intelligence(AI), this paper serves to examine its role in the corporate compliance world with an emphasis on hiring practices and policies. Current AI technology allows for companies to handle, sort, filter, and assess thousands of applications and resumes a day, all without the interference of humans. This shift from human to human hiring to AI to human is one that can be implemented, but must be done with caution. Ethical dilemmas such as bias, privacy, and lack of human interactions play major roles in determining if this process is for every business.
ContributorsValencia, Zacarias (Author) / Koretz, Lora (Thesis director) / Pofahl, Geoff (Committee member) / Barrett, The Honors College (Contributor) / Department of Finance (Contributor) / Dean, W.P. Carey School of Business (Contributor)
Created2024-05
Description
If quantum computing becomes feasible, many popular cryptographic schemes, such as RSA, Diffie-Helman, and methods using elliptic curves will no longer be secure. This paper explores code-based cryptography, specifically looking the McEliece cryptosystem, as well as the more recent Classical McEliece cryptosystem, which was proposed to the National Institute of Standards and Technology (NIST) as a potentially quantum-secure algorithm.
ContributorsFletcher, Julia (Author) / Sprung, Florian (Thesis director) / Childress, Nancy (Committee member) / Barrett, The Honors College (Contributor) / School of Social Transformation (Contributor) / School of Mathematical and Statistical Sciences (Contributor) / Historical, Philosophical & Religious Studies, Sch (Contributor)
Created2024-05
Description
Music festivals are often used as a way to generate money for a company or cause. In the case of Keep Tempe Beautiful, the organization is looking to create a sustainable music festival that will generate revenue to keep the organization going. Keep Tempe Beautiful is an organization that is volunteer-based with goals to end littering, improve recycling, and keep the city looking beautiful. Creating a sustainable festival, named Tempe City Roots, that aligns with these goals comes with research, problem-solving, and lots of planning.
As seen in the reflection and presentations, I have helped plan Tempe City Roots and created sustainable solutions to implement at the festival. Along with planning the overall festival (logistics, branding, agreements), I created my own brand, Friendly Festival, that follows the guidelines created for Tempe City Roots and will apply the guidelines to other festivals so they can implement and create a sustainable and environmentally friendly festival.
ContributorsCahill, Alexandra (Author) / Trujillo, Rhett (Thesis director) / Kuhn, Anthony (Committee member) / Barrett, The Honors College (Contributor) / Department of Marketing (Contributor) / Department of Finance (Contributor)
Created2024-05
DescriptionBuck-It is a budgeting application designed to meet the unique needs of college students. As financial literacy is crucial for developing good long-term financial habits, Buck-It aims to promote budgeting among college students through an appealing user interface, robust customization, and effective categorization.
ContributorsDoyle, Michael (Author) / Davitt, Ryan (Co-author) / Walle, Andrew (Co-author) / Vemuri, Rajeev (Co-author) / Baptista, Asher (Co-author) / Byrne, Jared (Thesis director) / Lee, Peggy (Committee member) / Barrett, The Honors College (Contributor) / School of Mathematical and Statistical Sciences (Contributor)
Created2024-05
DescriptionBuck-It is a budgeting application designed to meet the unique needs of college students. As financial literacy is crucial for developing good long-term financial habits, Buck-It aims to promote budgeting among college students through an appealing user interface, robust customization, and effective categorization.
ContributorsDavitt, Ryan (Author) / Doyle, Michael (Co-author) / Walle, Andrew (Co-author) / Vemuri, Rajeev (Co-author) / Baptista, Asher (Co-author) / Byrne, Jared (Thesis director) / Lee, Peggy (Committee member) / Barrett, The Honors College (Contributor) / Department of Economics (Contributor) / Department of Finance (Contributor)
Created2024-05
Description
This world promises just one thing: continuous change. As humanity has moved through time much has changed in the worlds of science, mathematics, and physics. These shifts in humanity's comprehension often arrive unexpectedly, driven by education, innovation, and experimentation. Such transformative waves resemble a series of technology shocks that are known to cause significant disruptions within an industry and the economy broadly as firms permanently change the ways they produce and distribute goods and services in response to new technologies or information. The recent flurry of innovation and interest in Artificial Intelligence leads us to believe that many industries may be experiencing such a wave of change today. The healthcare industry currently employs the most workers of any other sector in the United States (outside of the government) and is made up of an unprecedented 77% of female workers making the outcomes of changes in its labor market demands particularly important. In this paper we discuss the current state of Artificial Intelligence adoption within the clinical side of healthcare, what sub sectors and occupations are most exposed, and to what extent the FDA approved AI-enabled clinical healthcare products replace or complement those tasks of existing occupations. We also interviewed a few healthcare professionals with different levels of seniority and exposure to AI-enabled products to develop a holistic understanding of current AI adoption, employee preparation, and potential labor market implications over the short and long term. We find that AI implementation within clinical healthcare settings is young in its life cycle yet fast growing. Current use cases are mostly in the earlier stages of the patient’s care journey assisting workers in various capacities in the processes of patient testing, diagnosis, care planning, and post-treatment monitoring. The tasks associated with patient interaction and care administration do not appear to be threatened by AI automation at this point in time. Additionally, approved Artificial Intelligence products for clinical use are disproportionately concentrated in the subsectors of radiology, neurology, and cardiology. Finally, our interviews revealed a concerning lack of consideration and preparation, among healthcare workers, for the potential automation of their fundamental tasks. Going forward, we believe it wise for healthcare workers to monitor the evolution of clinical AI use cases as well as the FDA approval of AI-enabled products and prepare for potential automation by continuing to learn new skills, take on additional responsibilities, and generally inject themselves into as many stages of the patient’s healthcare journey as possible to differentiate among other workers and avoid the coming wave of mass clinical automation.
ContributorsMcElenney, Nicholas (Author) / Dolasinski, Nicholas (Co-author) / Asheim, Brody (Co-author) / Mehta, Ari (Co-author) / Calvo, Paula (Thesis director) / Don, Rachel (Committee member) / Barrett, The Honors College (Contributor) / Department of Finance (Contributor)
Created2024-05
Description
Concero Partners is committed to pioneering innovative capital solutions in emerging markets, with a strategic focus on Latin America. By leveraging a state-of-the-art online brokerage platform, Concero Partners aims to democratize access to global capital for small to mid-sized businesses in Chile and Brazil, enhancing financial transparency and fostering economic growth. At the heart of our mission is bridging the capital accessibility gap and simplifying transactions by collaborating with regional audit firms to standardize financial reporting, enhance market transparency, and make businesses more attractive to global investors. Beyond facilitating financial transactions, our broader goal is to stimulate economic growth by creating a more interconnected, efficient, and robust M&A ecosystem.
ContributorsSeykora, Colton (Author) / Freiheit, William (Co-author) / Harmon, William (Co-author) / Dalton, Ryan (Co-author) / Byrne, Jared (Thesis director) / Thomasson, Anna (Committee member) / Barrett, The Honors College (Contributor) / Historical, Philosophical & Religious Studies, Sch (Contributor) / Department of Finance (Contributor)
Created2024-05
Description
Concero Partners is committed to pioneering innovative capital solutions in emerging markets, with a strategic focus on Latin America. By leveraging a state-of-the-art online brokerage platform, Concero Partners aims to democratize access to global capital for small to mid-sized businesses in Chile and Brazil, enhancing financial transparency and fostering economic growth. At the heart of our mission is bridging the capital accessibility gap and simplifying transactions by collaborating with regional audit firms to standardize financial reporting, enhance market transparency, and make businesses more attractive to global investors. Beyond facilitating financial transactions, our broader goal is to stimulate economic growth by creating a more interconnected, efficient, and robust M&A ecosystem.
ContributorsDalton, Ryan (Author) / Freiheit, William (Co-author) / Seykora, Colton (Co-author) / Harmon, William (Co-author) / Byrne, Jared (Thesis director) / Thomasson, Anna (Committee member) / Barrett, The Honors College (Contributor) / Historical, Philosophical & Religious Studies, Sch (Contributor) / Department of Finance (Contributor)
Created2024-05
DescriptionBuck-It is a budgeting application designed to meet the unique needs of college students. As financial literacy is crucial for developing positive long-term financial habits, Buck-It aims to promote budgeting among college students through an appealing user interface, robust customization, and effective categorization.
ContributorsWalle, Andrew (Author) / Doyle, Michael (Co-author) / Davitt, Ryan (Co-author) / Vemuri , Rajeev (Co-author) / Baptista, Asher (Co-author) / Byrne, Jared (Thesis director) / Lee, Peggy (Committee member) / Barrett, The Honors College (Contributor) / Department of Finance (Contributor)
Created2024-05