{ Dan Grahn | AI Researcher. }

Research

MLsec

LLMs are increasingly capable of generating code, but this code is far from safe to deploy. How can we use AI to secure software that is written by human and mahcines?

AI Ethics

We are experiencing an epochal change in humanity's technological abilities. As we approach this new industrial revolution, how can we ethical and responsibly deploy AI to ensure we maximize benefits while minimizing costs?

Explainable AI & Trust

It's hard to understand the applied statistics of deep learning. It's even harder to trust those statistics with driving a car or flying a plane. How do we allow AI users to build trust in these systems?

Graph-Based Learning

Graphs are beautifully general data structures which can store any type of information, from binary call graphs to social networks. Traditional AI techniques aren't designed to operate on graphs. So how do we overcome this limitation?

Big Data Management

When you work with AI long enough, you'll encounter a data set that's too big for traditional tools. How we store, manage, and analyze that data is critical. So what is the best and most cost effective way to do that?

Publications

2023	Code Execution Capability as a Metric for Machine Learning&en;Assisted Software Vulnerability Detection. Daniel Grahn, Lingwei Chen, Junjie Zhang. International Symposium on Intelligent and Trustworthy Computing, Communications, and Networking. Conference Invited Assessing the Overlap of the DoD RAI Intiatives with the NIST AI RMF. Daniel Grahn. Military Operations Research Society 91st Symposium. Conference Panel Discussion on Data and Information Fusion Techniques to Enhance Digital Twin Applications. Daniel Grahn, et. al. SPIE Defense + Commercial Sensing, Signal Processing, Sensor/Information Fusion, and Target Recognition XXXII. Conference Invited
2022	The Prediction Management Framework: Ethical, Governable, and Interpretable Deployment of Artificial Intelligence / Machine Learning Systems. Daniel Grahn, Melonie Richey. SPIE Defense + Commercial Sensing (2022). Conference (+ best paper) Military Operations Research Society 90th Symposium, Data Science & Analytics Working Group Conference Military Operations Research Society 90th Symposium, AI & Autonomy Systems Working Group Conference
2021	An Analysis of C/C++ Datasets for Machine Learning-Assisted Software Vulnerability Detection. Daniel Grahn. Conference on Applied Machine Learning for Information Security (2021). Conference Military Operations Research Society 90th Symposium, Data Science & Analytics Working Group. Conference Military Operations Research Society 90th Symposium, Cyber Operations Working Group. Conference Satellite Dish Detection as a Semi-Supervised Small-Object Localization Problem. Daniel Grahn. Military Operations Research Society 89th Symposium (2021). Conference (+ best paper) mil-benchmarks: Standardized Evaluation of Deep Multiple-Instance Learning Techniques. Daniel Grahn. arXiv preprint arXiv:2105.01443 (2021). Preprint
2020 & Earlier	Potential impacts of elevated aerosol layers on high energy laser aerial defense engagements. Steven Fiorino, Stephen Shirey, Michelle Via, Daniel Grahn, Matthew Krizo. Atmospheric Propagation IX, vol. 8380, p. 83800T. International Society for Optics and Photonics (2012). Journal

Education

Ph.D.	Wright State University	Computer Science	2023
M.S.	University of Southern California	Computer Science	2018
B.S.	Cedarville University	Computer Science, Minor in Mathematics	2013

Experience

Axcel ILT	Director of Data Science & AI Training and Curriculum	2024–0000
Altamira Technologies	Principal AI Scientist	2020–2024
Northrop Grumman	AI Researcher, Principal Software Engineer	2013–2019
Wright State University	Adjunct Instructor	2019–2019
Air Force Institute of Technology	Adjunct Instructor	2018–2018
Air Force Institute of Technology	Research Assistant	2011–2013
Escape Plan Design	Web Applications Lead	2009–2011
Main-1-Media	Web Team Lead	2006–2009

Teaching Experience

CS-1160 - Introduction to Computer Programming	Wright State University	Fa21, Sp22, Fa22, Sp23
CEG-2170 - Introduction to C for Scientist and Engineers	Wright State University	Fa19
CS-589 - Operating Systems	Air Force Institute of Technology	Su18

My name is Dan Grahn.

And I research AI.