Brown Bag Seminar
The brown bag seminar is a weekly meeting organized by and for graduate students. The goal of the brown bag is to encourage students to practice presenting their work by giving talks to each other in a casual setting.
Hot topics in Solar Panels
DateFriday, October 5, 2018 - 12:00pm
AbstractThere's potential for solar energy to provide the totality of the world's energy consumption. But there are still barriers to overcome to achieve this potential. In this talk we will explore an overview of how solar energy is generated and managed through solar panels, what the state of the art is, and what are the current market interests and research trends for improving technical, infrastructure, economic and policy barriers in upcoming years. We will look in particular at bifacial solar panel technology, and how industry and researchers alike are contributing to make this low-risk high-gain technology the next game-changer in photovoltaics. Bio: Silvana Ayala Pelaez is a PhD candidate in Electrical and Computer Engineering at the University of Arizona. She has an M.S. in Optical Sciences at the same University. She received a B.S. in Mechatronics Engineering from Monterrey Tec (ITESM 2007). She works in the Photonic Systems Laboratory under Dr. Raymond Kostuk. Her primary area of expertise is on the integration of holographic optical components with photovoltaic materials for higher efficiency in solar panels. Current projects are focused on bifacial photovoltaic performance and modeling. Her research includes characterization and energy simulation for bifacial and bifacial/holographic system energy productions. She has two papers under review in collaboration with the National Renewable Energy Laboratory, in Golden, Colorado. She edited and published the book “Solar Outreach Handbook” in 2018.
MCMC for High Energy X-Ray Radiography
DateFriday, October 12, 2018 - 12:30pm
AbstractImage deblurring via deconvolution can be formulated as a hierarchical Bayesian inverse problem, and numerically solved by Markov Chain Monte Carlo (MCMC) methods. Numerical solution is difficult because inconsistent assumptions about the data outside of the field of view of the image lead to artifacts near the boundary; and the Bayesian inverse problem is high-dimensional for high-resolution images. The numerical MCMC framework I present addresses these issues. Boundary artifacts are reduced by reconstructing the image outside the field of view. Numerical difficulties that arise from high-dimensions are mitigated by exploiting sparse problem structure in the prior precision matrix.
Optimization Reformulations and Algorithms in Machine Learning
DateFriday, October 19, 2018 - 12:00pm
AbstractAll training of machine learning models can be represented as an optimization program. Thus training the model is actually finding the optimal solution to a typically non-linear program. It is known in optimization that there are many equivalent formulations of the optimization program and that which one is better depends on the algorithm used. We will use a specific multiclass support vector machine (SVM) model called Multicategory SVM which extends the interpretation and statistical properties of binary SVM. We will show the difficulties associated with this formulations. We will consider two optimization algorithms: coordinate descent and projected subgradient methods. We will find the best formulation of MC SVM for each method.