Deep Purple: Deep Purposeful Learning

Deep Purple aims to advance the modeling of complex dynamic systems using new information-efficient approaches that make optimal use of data and known physics at multiple scales. The program is investigating next-generation deep learning approaches that use not only high throughput multimodal scientific data from observations and controlled experiments (including behaviors such as phase transitions and chaos), but also of the known science of such systems at whatever scales it exists.

Using data sets from the biological and physical sciences, the program is developing: 1) new methods for de-noising and interpolating stochastic time-series data; 2) generative models for predicting systems trajectories, resilience, and stability; and 3) new approaches to modulate final state trajectories of such systems.