digiLab Newsletter: 1

If you want to receive future newsletters straight to your inbox, signup here!

An ML workshop, in collaboration with Tokamak Energy:

We focused on the application of mission-critical ML to fusion energy development workflows. Attendees discovered how to use our ML platform, twinLab, to combine their fusion expertise with the latest in probabilistic modelling and uncertainty quantification (UQ).

Read More

twinLab 2.0 released:

twinLab also got some powerful updates at the start of the year! Our product team have:

1. Revamped the Interface - we've made the Python client more intuitive by separating twinLab's functionality into clearer modules. This is designed to increase the ease and time-to-value for users solving complex data challenges.

2. Introduced Functional Decomposition - this enables you to train a probabilistic ML model on datasets with high dimensionality (lots of input/feature columns). twinLab can now do this automatically, to accelerate the model training speed, while minimising any loss in accuracy.

3. Improved Uncertainty Quantification - twinLab also now enables you to understand exactly where uncertainty exists in your data, even if it's intrinsically complex and correlated.

4. Enabled the Recommend Module - twinLab Recommend tells you where to run your next experiment, or where to take your next sample. This reduces the amount of samples/experiments required to understand your parameter space, saving you time and/or money.

Explore twinLab

Fusion Industry News

Dr Cyd Cowley, who recently joined us as a Fusion Solutions Consultant, released another episode of Fusion News for the Fusion Industry Association, wearing his digiLab merch for the first time!

He explored opinions from the US and Germany about power plant funding, the interaction between Fusion and AI at the World Economic Forum, and a futuristic project to clear space debris with lasers from EX-Fusion Inc.

Watch the Video

What is twinLab?

twinLab is our cloud platform for applying probabilistic Machine Learning to your simulations, experiments, or sensor data. This means it adds Uncertainty Quantification to your model outputs, so it's especially useful when you need to make predictive decisions based on limited or sparse data.

Functionality is exposed through a rich and expressive RESTful API, and we've built a lightweight Python client as our first interface. You can learn more about twinLab's capabilities on our website.

Trying twinLab

twinLab is already being used to solve next-generation engineering and sustainability challenges for organisations like the UKAEA, Airbus, and Rolls-Royce. Our mission is to help more people and organisations use probabilistic ML.

Just hit "Try twinLab" and we'll set you up with an API key, the documentation, and some example solutions.

Alternatively, if you've got a particular data challenge, book a free call with one of our solution engineers.