--[ Rapid Prototyping of Machine Learning Solutions

$ getent passwd mcruickshank
├─── name: Major Iain J. Cruickshank
├──── org: ACI
└─ social:
   └─ linkedin: linkedin

Major Iain Cruickshank is a research scientist at the Army Cyber Institute, where he researches computational social science methodologies and machine learning techniques. He has previous assignments at the 780th MI BDE and the Army's Artificial Intelligence Integration Center. He is also an active competitive data scientist with notable wins in tabular, computer vision, and text-based competitions.


Have you ever come across a task where you think a machine learning (ML) model might help automate or even do the task better and faster than you can? Have you ever wanted to experiment with implementing your own ML solutions, but aren’t sure where to start? In this tutorial, I will be showing you how to implement quick machine learning solutions, from open source and freely available tools, for a variety of real-world problems. As part of the tutorial, we will explore the methodology for implementing quick, real-world ML solutions across a variety of data scenarios (i.e. text, tabular, image). We will then get hands-on with the actual code and processes for implementing these solutions. Finally, we conclude the tutorial by doing a timed competition to implement a machine learning model on real data. At the conclusion of the tutorial, participants will have gained hands-on experience with implementing state-of-the-art machine learning solutions to real-world problems with various types of data.


For the tutorial, students will need to have at least a basic level of proficiency with Python; understand functions, objects, loops, control flows, and variables. Knowledge of data-centric Python (i.e. Pandas, file I/O, etc.) is preferable.