We conduct the basic machine learning research needed to estimate representations of biomedical data that are
- Robust
- Interpretable
- Data efficient
- Reflective of inherent data uncertainty
- Able to leverage existing knowledge
These representations are both predictive and knowledge discovery tasks.
Our research focuses on four themes, and each theme advances different aspects of representation learning for life science and support each other:
- Meaningful representation of data and computational and mathematical tools development to realize the answer.
- Geometric constructions to incorporate existing knowledge into representations and ensure that the result is understandable by humans.
- Representation of data often appearing within life science, such as trees, graphs, and sequences.
- Inclusion of real data that is “noisy” and investigation of how associated uncertainty is best encoded.