Looking for ways of presenting knots which help artificial intelligence to learn to manipulate knots

  • Thu 28 Oct 21

    15:00 - 16:00

  • Online

    Zoom (email for link)

  • Event speaker

    Alexei Vernitski

  • Event type

    Lectures, talks and seminars

  • Event organiser

    Mathematical Sciences, Department of

  • Contact details

    Jesus Martinez-Garcia

These Departmental Seminars are for everyone in Maths. We encourage anyone interested in the subject in general, or in the particular subject of the seminar, to come along. It's a great opportunity to meet people in the Maths Department and join in with our community.

Looking for ways of presenting knots which help artificial intelligence to learn to manipulate knots

Knots (like the one presented in the picture) are difficult to begin to study mathematically because mathematical notation works well with words or matrices, and a knot diagram cannot be easily represented as either. This is why in knot theory much effort is invested in representing knots in the form of words or matrices (for example, you might have heard of Gauss words or Goeritz matrices).

Now suppose we want the computer to work with knots; then we face a different kind of problem, namely, the computer does not possess human 2D and 3D intuition. To enable the computer to start exploring knots, we need to trawl through existing representations of knots (or invent new ones) looking for those which will compensate for the computer not possessing human spatial intuition.


Alexei Vernitski, University of Essex

How to attend

If not a member of the Dept. Mathematical Science at the University of Essex, you can register your interest in attending the seminar and request the Zoom’s meeting password by emailing Dr Jesus Martinez-Garcia (jesus.martinez-garcia@essex.ac.uk)

A thick piece of twisted brown rope tied in to a large knot.
Project: Machine learning for recognising tangled 3-D objects

This project, funded by the Leverhulme Trust, will help teach computers how to understand 3-dimensional space without using deep neural networks.

Related events