CAIRE develops automated machine-learning (AutoML) methods that are accessible to everyone. Center activities include the development of fair and unbiased AutoML, knowledge-guided AutoML, education and training, and collaboration to apply AutoML to biomedical research, precision medicine and improving healthcare.
Automated Machine Learning
Jason Moore, PhD, leads a team focused on developing novel automated approaches for machine learning (AutoML) with the goal of making AI accessible to everyone. His team developed the Tree-based Pipeline Optimization Tool (TPOT), which was one of the very first AutoML software packages to be released. This was followed by the user-friendly Aliro software, which makes machine learning accessible to even the most novice user. His current focus is on tailoring these methods to biomedical and clinical data and integrating them with biomedical knowledge bases for informed machine learning.
Fairness and Bias
Tiffani Bright, PhD, leads a team focused on developing algorithms and methods for detecting and correcting biases in machine-learning results to ensure models deployed in the clinic treat all patients equally. One of her particular interests is extending AutoML methods and software such as Aliro and TPOT to the automated detection and correction of biases to produce fair models for clinical application.
Have Questions or Need Help?
Contact us if you have any questions or wish to learn more about the programs and services at the Center for Artificial Intelligence Research and Education.