Skip to content

June Meetup: Rediscovering orbital mechanics with machine learning

Photo of Peter Hurley
Hosted By
Peter H.
June Meetup: Rediscovering orbital mechanics with machine learning

Details

Dear All,
I am pleased to invite you to June's Sussex Data Science Meetup, our first hybrid event, both on zoom, and in-person, at the Spaces working space in Mocatta House, Trafalgar Place

In-person attendees: please arrive by 18:30 for snacks and socializing before the talk and make sure you are registered to attend so we have an idea of numbers for catering.
Remote attendees: we will start broadcasting the meeting at 19:00 at this zoom link, there is no need to register

This month's talk is being presented by Dr Pablo Lemos, a postdoctoral Research Fellow in cosmology at University of Sussex with a keen interest in machine learning.

Pablo will be talking about his recent project: Rediscovering orbital mechanics with machine learning

We present an approach for using machine learning to automatically discover the governing equations and hidden properties of real physical systems from observations. We train a "graph neural network" to simulate the dynamics of our solar system's Sun, planets, and large moons from 30 years of trajectory data. We then use symbolic regression to discover an analytical expression for the force law implicitly learned by the neural network, which our results showed is equivalent to Newton's law of gravitation. The key assumptions that were required were translational and rotational equivariance, and Newton's second and third laws of motion. Our approach correctly discovered the form of the symbolic force law. Furthermore, our approach did not require any assumptions about the masses of planets and moons or physical constants. They, too, were accurately inferred through our methods. Though, of course, the classical law of gravitation has been known since Isaac Newton, our result serves as a validation that our method can discover unknown laws and hidden properties from observed data. More broadly this work represents a key step toward realizing the potential of machine learning for accelerating scientific discovery.

Sussex Data Science Meetups are supported by:

  1. DISCUS (http://www.sussex.ac.uk/discus/)
  2. Silicon Brighton (http://www.siliconbrighton.com)
  3. Mojeek (https://www.mojeek.com/).

We are still looking for a mix of talks from a variety of backgrounds/subjects/speakers.
Please contact me if you are interested in being a speaker at future meetups: p(dot)d(dot)hurley(at)sussex(dot)ac(dot)uk

COVID-19 safety measures

Event will be indoors
The event host is instituting the above safety measures for this event. Meetup is not responsible for ensuring, and will not independently verify, that these precautions are followed.
Photo of Sussex Data Science group
Sussex Data Science
See more events
Spaces - Brighton, Trafalgar Place
Trafalgar Place · Brighton