Learn Data Skills
Katelyn S.

Katelyn S.

Researcher

UPenn

Technologies

My Portfolio Highlights

My New Article

Determining the Optical Depth to Reionization using 21 cm Data and Robust, Generalizable Machine Learning Simulating the Ionization Fraction: Understanding the Early Universe with CNN Models

My New Workbook

Project: Investigating Netflix Movies

My New Article

Determining the Optical Depth to Reionization using 21 cm Data and Robust, Generalizable Machine Learning Simulating the Ionization Fraction: Understanding the Early Universe with CNN Models

Insights architect, constructing bridges between data and actionable strategies.

My Work

Take a look at my latest work.

DataLab

Project: Investigating Netflix Movies

DataLab

Project: Analyzing Students' Mental Health

article

Determining the Optical Depth to Reionization using 21 cm Data and Robust, Generalizable Machine Learning Simulating the Ionization Fraction: Understanding the Early Universe with CNN Models

DataLab

Project: Investigating Netflix Movies

My Certifications

These are the industry credentials that I’ve earned.

Other Certificates

NVIDIA Fundamentals of Deep Learning

DataCamp Course Completion

Take a look at all the courses I’ve completed on DataCamp.

Courses

Introduction to Python
Introduction to SQL
Intermediate Python
Intermediate SQL

My Work Experience

Where I've interned and worked during my career.

University of Pennsylvania | Aug 2023 - Jan 2024

Post-Bacc Researcher

Currently, using machine learning techniques to understand the Epoch of Reionization (EoR) using 21cm observations that appear in the Cosmic Microwave Background. To estimate the ionization fraction, I am tasked to create a convolutional neural network (CNN) with both convolutional and dense layers and train it using simulated 21 cm data. We present results on how well the networks generalized by considering the variance and bias as a function of the ionization fraction, and present recommendations for training future networks to be robust against the differing physical assumptions of simulations. In this work, we use simulations from the 21cmfast and zreion simulator packages.
Show More

Texas Christian University | Aug 2023 - Dec 2023

Undergraduate Student Researcher

Analyzed the Sagittarius dwarf galaxy core and its chemical gradient using the DR16 dataset release that was trained using The Cannon learning system.

Simons Foundation | May 2023 - Aug 2023

Simons-NSBP Scholar

In this project, I investigated the overlap between Simons Observatory observing regions and other experiments which will have useful data on the Epoch of Reionization (EoR) (e.g, SPHEREx, Roman, HERA, SKA). This overlap requires understanding the EoR, the celestial coordinate system, and being able to do research to find the relevant survey footprints and plot them. The purpose of examining these near-future missions will allow us to understand the overlap of the observations on the sky and the complementary nature of the physical probes they provide for reionization. Beyond my investigation, this will include looking in more detail at the data that will soon be available for each observatory and how best it might be combined with SO data to understand the EoR, complementary to SO science goals. Additionally, I am understanding how to train networks in simulating theoretical observations of reionization using machine learning techniques.

Texas Christian University | Feb 2021 - Dec 2022

Undergraduate Student Researcher

The first galaxies formed 12.5 billion years ago during the first billion years after the Big Bang. However, these first, faint, galaxies remain too faint for direct detection, even by our most powerful telescopes. Therefore we study them using their fossils relics, ultra-faint dwarf galaxies orbiting the Milky Way. In this work, I look at the histories of star formation in simulated analogs to the ultra-faint dwarfs. These star formation histories will allow us to study the details of how and when star formation occurred during the first billion years of cosmic time. I was particularly interested in how massive the first galaxies were when they formed the majority of their stars.

Texas Christian University | Sep 2021 - Dec 2021

Observational Researcher

Conducted observational research at the McDonald Observatory in Fort Davis, Texas. Used a ST-16803 CCD to capture images of the globular cluster on the 0.9 m and 2.1 m telescopes with fellow classmates. This observational run was for three nights. After collecting data, we are transferring the information over to Jupyter notebook to further reduce data and compute calculations.

My Education

Take a look at my formal education

Bachelor of Science in Astronomy & PhysicsTexas Christian University | 2024

About Me

Katelyn S.

A recent Astrophysics graduate who is looking to broaden their skills in the tech industry. My experience is grounded in cosmology and stellar astrophysics research utilizing the Python language.

Powered by

  • Work
  • Certifications
  • Courses
  • Experience
  • Education
  • About Me
  • Create Your Data Portfolio for Free