We use cookies to enhance your experience. By continuing to browse this site you agree to our use of cookies. More info.

When we were at school we probably believed that science and mathematics were primarily concerned with offering definitive answers. But grown-up scientists and engineers know that this is rarely the case.  Pa6 Thermoplastic

Uncertainty is inherent in almost every system studied and every experiment conducted. It might come from natural scatter in the experimental results, limitations in the precision of measurement methods, errors made when performing measurements, or the need to extrapolate from incomplete data.

We also need to deal with uncertainty in machine learning (ML). Predictions from ML models will be uncertain partly due to inherent scatter in the training data, and partly because models are built from limited data.

It is important to embrace this uncertainty, rather than expecting to eliminate it entirely. It needs to be understood, quantified, and used to make better decisions.

In fact, uncertainty is a positive. Just as a straight line that passes through all points on a graph reveals how straightforward a system is to characterize, a model containing negligible uncertainty is not likely to be very useful or insightful. It is the difficult and complex problems that come with uncertainty, and these are the issues that are valuable to resolve.

ML specialist Intellegens has taken every care to ensure that uncertainty is faithfully quantified within the predictions of its Alchemite™ algorithm. The algorithm expands on conventional methods that, at worst, do not really estimate uncertainty at all but assume a normal distribution, or only capture one of many contributions to the overall uncertainty.

Accurate quantification makes it possible, for example, to identify the likelihood of an experiment succeeding so that the user can target experimental work accordingly.

Intellegens CSO, Dr. Gareth Conduit, introduced some interesting new research work from his group at the University of Cambridge in a recent webinar. The work performed by Dr. Conduit and his team goes a little further in embracing uncertainty.

It focused on the design of concrete, for which measurements of many parameters associated with the material appear ‘noisy.’

Rather than simply quantifying the uncertainty, this work allowed the machine learning method to learn from the noise in the data, discovering that the uncertainty in one particular physical parameter could be used to help predict the concrete strength.

Two new concrete mixes were proposed as a result of this work, subsequently created, and then found to behave as predicted, which expanded the properties of mixes that are commercially available.

If we’re looking for some certainty in all of this it is that machine learning can do interesting things in the world of chemicals and materials. But only if you take the right approach to uncertainty!

This information has been sourced, reviewed and adapted from materials provided by Intellegens Limited.

For more information on this source, please visit Intellegens Limited.

Please use one of the following formats to cite this article in your essay, paper or report:

Intellegens Limited. (2022, October 20). Embracing the Uncertainty in Modeling with Machine Learning. AZoM. Retrieved on October 27, 2022 from https://www.azom.com/article.aspx?ArticleID=22018.

Intellegens Limited. "Embracing the Uncertainty in Modeling with Machine Learning". AZoM. 27 October 2022. <https://www.azom.com/article.aspx?ArticleID=22018>.

Intellegens Limited. "Embracing the Uncertainty in Modeling with Machine Learning". AZoM. https://www.azom.com/article.aspx?ArticleID=22018. (accessed October 27, 2022).

Intellegens Limited. 2022. Embracing the Uncertainty in Modeling with Machine Learning. AZoM, viewed 27 October 2022, https://www.azom.com/article.aspx?ArticleID=22018.

Do you have a review, update or anything you would like to add to this article?

This interview discusses how ELTRA, as part of Verder Scientific, produces elemental analyzers for the battery component space.

James Stephenson & Selina Ambrose

After connecting at the Advanced Materials Show 2022, AZoM spoke with Selina Ambrose and James Stephenson from Promethean Particles about the mission of the company and the way MOFs are changing the game of carbon capture.

This interview with Francis Quintal Lauzon, Director of Product and Innovation at Clemex Technologies, discusses using artificial intelligence in image analysis. He discusses the challenges involved with automation in microscopy, as well as possibilities for the future of AI.

The BitUVisc is a unique model of viscometer, that can manage highly viscous samples. It is designed to keep the samples warm throughout the entire process.

Discover the FieldLab analyzer, a new, battery-powered, portable oil lab by Spectro Scientific that allows for rapid oil testing in the field.

The BeVision D2 is a dynamic image analyzer that is the perfect choice for the dynamic image analysis of dry powder and granules with its particle range of 30-10,000 microns and ability to analyze 24 different parameters.

This article provides an end-of-life assessment of lithium-ion batteries, focusing on the recycling of an ever-growing amount of spent Li-Ion batteries in order to work toward a sustainable and circular approach to battery use and reuse.

Corrosion is the degradation of an alloy caused by its exposure to the environment. Corrosion deterioration of metallic alloys exposed to the atmosphere or other adverse conditions is prevented using a variety of techniques.

Due to the ever-increasing demand for energy, the demand for nuclear fuel has also increased, which has further created a significant increase in the requirement for post-irradiation examination (PIE) techniques.

AZoM.com - An AZoNetwork Site

Thermoplastic Composites Examples Owned and operated by AZoNetwork, © 2000-2022