Computational Discovery of Scientific Process Models

Abstract

The PIs, Pat Langley and Charbel Farhat, provides a compelling research agenda for developing a machine learning (ML) framework for scientific discovery. The PIs identify a number of key aspects that such a framework needs to incorporate- causal modeling, human interpretability, applicability in moderate data settings, and an effective closed-loop setting, where model formulation, reformulation and experimentation are interleaved in a continuous loop. These ML methods can find useful statistical regularities in large data sets and can be used to construct predictive models, but these models lack an interpretable structure and do not fit into numerous existing scientific theories that have been shown to be so remarkably effective. To address the shortcomings of existing approaches, the PIs proposes to develop several novel extensions of the current methods for inductive process modeling.

Document Details

Document Type
DoD Grant Award
Publication Date
Mar 07, 2024
Source ID
FA95502310580

Entities

People

  • Pat Langley

Organizations

  • Air Force Office of Scientific Research
  • United States Air Force

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks