Build the future: Design concrete with machine learning

Add to your list(s) Download to your calendar using vCal

• Bogdan Zviazhynski, TCM Group, Cavendish Laboratory
• Thursday 16 November 2023, 15:00-16:00
• Mott Seminar Room.

If you have a question about this talk, please contact Stephen Walley.

Noise is usually the enemy of machine learning as uncertainty in training data leads to inaccurate predictions. However, we develop and validate a machine learning architecture that extracts crucial information out of the noise itself to improve the predictions. We apply the formalism to specify a concrete mix that has low environmental impact, and another concrete mix that has high resistance to carbonation, and both mixes also fulfil targets on the strength, density, and cost. The proposed concretes were then mixed, set, and the predicted properties experimentally verified. Our generic formalism enables the exploitation of uncertainty in machine learning, which has a broad range of applications in the physical sciences and beyond.

This talk is part of the Physics and Chemistry of Solids Group series.

This talk is included in these lists:

• All Cavendish Laboratory Seminars
• All Talks (aka the CURE list)
• Centre for Health Leadership and Enterprise
• Featured lists
• ME Seminar
• Mott Seminar Room
• Neurons, Fake News, DNA and your iPhone: The Mathematics of Information
• Physics and Chemistry of Solids Group
• School of Physical Sciences
• Thin Film Magnetic Talks

Note that ex-directory lists are not shown.