Alexander L. Hayes is a Health Informatics Ph.D. Student working with Professor Kay Connelly and the Proactive Health Informatics Department on the Precision Health Initiative. His interests are in statistical relational artificial intelligence (STARAI), systems, open source development—and their applications toward solving real-world Health Informatics problems. He holds a B.S. Computer Science from Indiana University, and continues to collaborate with Professor Sriraam Natarajan and his colleagues from the StARLinG Lab at the University of Texas at Dallas.
He is currently working in the nuMoM2b (new-mom-2-be) data set, where one of the aims is to model the development of gestational diabetes. Additional information about this project is included in the nuMoM2b preprocessing documentation.
srlearn: A Python Library for Gradient-Boosted Statistical Relational Models
srlearn is a Python library for learning gradient-boosted statistical relational models.
- srlearn: A Python Library for Gradient-Boosted Statistical Relational Models was accepted at the Ninth International Workshop on Statistical Relational AI
- 2019-06-14: I am attending International Conference of Machine Learning (ICML) in Long Beach, California.
- Spring 2019: I am transferring to work as a Research Assistant with the Proactive Health Informatics (ProHealth) Department within the School of Informatics, Computing, and Engineering (SICE) at Indiana University, Bloomington.
- Fall 2018: I will be a Teaching Assistant for Automata Theory (CS 4384.001)
Recent blog posts
This site can't be reached? 127.0.0.1 refused to connect? How does Jekyll development on Chromebook differ from developing on other systems?
We rarely notice that we live inside an invisible layer of liquid heat. This describes how I glimpsed into the heat distribution of the IU Informatics Building.
k-means clustering can be used on images to automatically perform color quantization. This demonstrates how to cluster colors in an animation.
k-Nearest Neighbors sometimes gets a bad reputation for being too simple. So I implemented it in as few lines as possible.