Alexander D. Greenhalgh
PhD Student in Computer Science at Georgia Tech
I am a PhD student in Computer Science and Engineering at the Georgia Institute of Technology, advised by Dr. Polo Chau, where I study machine learning, scientific computing, and operations research.
Previously, I worked as an Operations Research Analyst at Army Futures Command, building combat simulation models and LLM-based tools for tactical decision support. I have worked with researchers and engineers at Georgia Tech, the Department of Defense, Oak Ridge National Lab, and the University of Tennessee.
News
|
ResearchMy work spans machine learning, scientific computing, and operations research. I am particularly interested in applying modern ML methods—including deep learning, optimization, and LLMs—to problems in materials informatics, autonomous systems, and decision support. Prior work includes developing interatomic potentials for high-entropy alloys via ML optimization, building retrieval-augmented LLM tools for military applications, and constructing predictive models for alloy mechanical properties. |
Publications |
Projects |
Designed and built a retrieval-augmented LLM chatbot in Python for tactical question-and-answer, using LangChain for vector embeddings and the OpenAI API. Training corpus consisted of Army Doctrine Manuals scraped with BeautifulSoup.
XGBoost model for joint optimization of yield strength and ductility, trained on an experimental High-Entropy Alloy database. Achieved 90.1% testing accuracy.
Implemented LSTM, Seq2Seq, and Transformer architectures in PyTorch for English-to-German translation. Reproduced attention from Vaswani et al.'s Attention Is All You Need. Testing perplexity of 4.81.
Implemented A*, PID control, Particle and Kalman Filters, and Simultaneous Localization and Mapping (SLAM).
|
Built with Jekyll |