Machine Learning and Decision Making Group

The Machine Learning and Decision Making Group develops new methods and tools for solving complex learning and decision-making problems under conditions of uncertainty. They apply these methods to applications in disease outbreak surveillance, treatment error detection, high throughput genomic and proteomic data analysis, the monitoring and learning of traffic flows, diagnosis, and strategic financial planning within organizations.

Within the Machine Learning Group, students conduct research in diverse fields. Several students conduct research in the bioinformatics domain, covering analyses in genomic and proteomic data, as well as mining novel concepts from biomedical literature. A major effort has also led to the development of clinical systems for diagnosis and anomaly detection, using the latest methods in temporal reasoning and feature extraction. In addition, methods have been developed for modeling complex traffic patterns and planning in reward-based environments.

The group also includes the Decision Systems Laboratory (DSL), which maintains a research and teaching environment for faculty and students interested in developing techniques and systems that support decision making under uncertainty as well as system modeling and discovery. Their methods include theoretical work in machine learning, Bayesian analysis and causal discovery, system building, and empirical studies. DSL relies on probabilistic, decision-theoretic, and econometric techniques combined with artificial intelligence approaches. DSL is the developer of GeNIe and SMILE, popular software for decision-theoretic modeling and learning, available on the BayesFusion website.

In addition, the group includes the ULab, an academic research team primarily focused on usability engineering. The ULab team is involved in a multidisciplinary program of research, including human-robotic interaction, human agent teams, group decision modeling, information fusion, virtual reality, human modeling using machine learning, information visualization, metaphor-based interaction and human factors.

Primary Faculty