The Science of Learning from Observations: Leveraging Scientific Computation with Intrinsic Machine Learning Models and Lifelong Learning

Abstract

Complex physical systems exhibit nonlinear, multi-scale, high-dimensional dynamics that challenge current methods of modeling, simulation and interpretability. We will study these issues with ideas and techniques in machine learning, both new and existing but not currently used in this context. We will introduce novel machine learning algorithms that synthesize (and, at times, invent new) fundamental structures in complex dynamical systems from partial observations and different sensor modalities, integrate physical constraints and boundary conditions, accumulate knowledge over observations of multiple dynamical systems and transfer it appropriately to new systems, and link them with computational technology

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 21, 2022
Source ID
FA95502110317XX0

Entities

People

  • Mauro Maggioni

Organizations

  • Air Force Office of Scientific Research
  • Johns Hopkins University
  • United States Air Force

Tags

Fields of Study

  • Computer science

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Neural Network Machine Learning.
  • Research Science/Academic Research

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks