Chase Gaudet

University of Louisiana at Lafayette | ULL · Department of Physics

We show that the core reasons that complex and hypercomplex valued neural networks offer improvements over their real-valued counterparts is the weight sharing mechanism and treating multidimensional data as a single entity. Their algebra linearly combines the dimensions, making each dimension related to the others. However, both are constrained to...

The field of deep learning has seen significant advancement in recent years. However, much of the existing work has been focused on real-valued numbers. Recent work has shown that a deep learning system using the complex numbers can be deeper for a fixed parameter budget compared to its real-valued counterpart. In this work, we explore the benefits...

Convolutional Neural Networks are extremely powerful visual models. We show that these models can be applied to object detection in point cloud data without engineered features. Our key insight is to rasterize the point cloud by way of different statistics and combining these statistics into a multi-channel 'image' we call a Statistical Raster or S...

A large amount of papers are uploaded to Cornell’s arxiv.org everyday. We attempt to give researchers an easier time navigating this massive open source database by creating a website of analysis tools based on the keywords of papers. The data is obtained from the arxiv’s API and run through a neural network to extract keywords. This method of usin...

It is an exciting time in the direct detection of dark matter. Many experiments are showing a signal, but most of them do not overlap. Our goal was to construct a simple non-relativistic effective field theory of dark matter-nuclei scattering. This is an extension on work done by Fan, Reece, and Wang where we do not take limiting cases on mediator...