AI for Science

Using AI/ML to do moar Sciencing!

What could be more important that using AI to learn how to do science more effectively, to learn what doing science really means, to update good methods for modelling our universe and make great?

In my lab we have done work on a few focussed topics in this area:

  • Combustion Modelling - We used Deep Learning to greatly improve the speed and scale possible for existing flamelet estimation models.
  • Material Design - In this ongoing work with the National Research Council, we are investigating exciting ways to apply Reinforcement Learning to the problem of material design and digital chemistry.

Our Papers on AI for Science

  1. Balancing Information with Observation Costs in Deep Reinforcement Learning
    Colin Bellinger, Andriy Drozdyuk, Mark Crowley, and Isaac Tamblyn.
    In Canadian Conference on Artificial Intelligence Canadian Artificial Intelligence Association (CAIAC), Toronto, Ontario, Canada, may, 2022.
  2. Scientific Discovery and the Cost of Measurement – Balancing Information and Cost in Reinforcement Learning
    Colin Bellinger, Andriy Drozdyuk, Mark Crowley, and Isaac Tamblyn.
    In 1st Annual AAAI Workshop on AI to Accelerate Science and Engineering (AI2ASE) feb, 2022.
  3. Amrl
    Active Measure Reinforcement Learning for Observation Cost Minimization: A framework for minimizing measurement costs in reinforcement learning
    Colin Bellinger, Rory Coles, Mark Crowley, and Isaac Tamblyn.
    In Canadian Conference on Artificial Intelligence Springer, 2021.
  4. Reinforcement Learning in a Physics-Inspired Semi-Markov Environment
    Colin Bellinger, Rory Coles, Mark Crowley, and Isaac Tamblyn.
    In Canadian Conference on Artificial Intelligence Springer, Ottawa, Canada (virtual), may, 2020.
  5. ECML
    Compact Representation of a Multi-dimensional Combustion Manifold Using Deep Neural Networks
    Sushrut Bhalla, Matthew Yao, Jean-Pierre Hickey, and Mark Crowley
    In European Conference on Machine Learning (ECML-19) Wurzburg, Germany, 2019.