The Massachusetts Institute of Technology’s Plasma Science and Fusion Centre (PSFC) held a collaborative workshop with digiLab, a machine learning (ML) company, to explore the advantages of probabilistic machine learning in advancing the development of fusion power.
The digiLab Team
On the 20th of March, delegates from digiLab attended the fusion industry association conference in Washington D.C.
The digiLab Team
digiLab recently co-hosted an ML workshop at the prestigious Royce Discovery Centre, bringing together industry and academia to explore the intersection of machine learning (ML) with the domains of forging, advanced materials, and manufacturing. The event drew representatives from key institutions including the Henry Royce Institute, Rolls Royce, the University of Strathclyde, the University of Manchester, WH Tildesley Ltd, and the University of Sheffield.
The digiLab Team
We've been travelling the globe for the last two weeks, attending the Fusion Industry Association annual policy conference in D.C., and delivering twinLab workshops for MIT, for The Henry Royce Institute / Rolls-Royce, and, nearer to home, for Exeter College. Cyd Cowley, fusion solutions engineer at digiLab, has released Part 1 of his "Machine Learning for Fusion" series. Meanwhile, Ana has published the final part of her Gaussian Processes series, finishing up with a look at "autoML" in twinLab. Finally, we're excited to announce our three summer interns.
The digiLab Team
We are delighted to introduce the students who will be taking part in digiLab’s 2024 Summer Internship Programme. Joe Rayson is a 2nd year Business and Economics student at the University of Exeter and will work in Business Development at digiLab. Natalia Wojcik is studying for a BSc in Economics at the University of Exeter. Lucas Pigott comes to digiLab fresh from a study abroad year at the University of North Carolina.
The digiLab Team
This fortnight, our Technical Content Lead, Ana, has finished the second video in her "Demystifying Gaussian Processes" series. She'll explore how we can choose the right kernel for our Gaussian Process. The twinLab team has also implemented Initial Design Spaces, to help users build models or experiment designs from scratch. Finally, we've joined the U.S. DIII-D National Fusion Facility!
The digiLab Team
digiLab has begun efforts to advance fusion in the US, by joining the DIII-D National Fusion Facility, a U.S Department of Energy user facility. As a member of this facility that houses a world-class research tokamak, digiLab can now work with data and scientists from the facility to solve some of the hardest challenges in fusion.
The digiLab Team
The big news this fortnight has been success in the water sector. This validates our goal of enabling organisations in multiple industries to take the next step in their digital transformation journey by deploying ML solutions. We've also augmented twinLab with dimensionality reduction capabilities.
The digiLab Team
We’ve received almost half a million pounds from OfWat to further develop and commercialise senSiteUQ, our sensor placement optimisation solution for water networks. senSiteUQ enables operators to maximise their understanding of the network, while minimising the total number of sensors required. That translates into better decision-making with reduced costs. senSiteUQ will enable the water sector to reduce CSO pollution, respond faster to extreme weather, improve biodiversity, and have improved live monitoring and data analysis.
The digiLab Team
It's been a busy two weeks in the fusion sector: we attended the 2024 FARSCAPE workshop at the UKAEA; and we're excited to announce the formation of our specialist fusion team, supported by key hiring. Meanwhile, our Technical Content Lead, Ana, has released the first part of her explainer video series demystifying Gaussian Processes!
The digiLab Team
During the workshop, we demonstrated our large language model for accessing fusion information. We also discussed ongoing and future projects in the programme from simulating plasma turbulence to providing uncertainty quantification for engineering systems!
The digiLab Team
digiLab is excited to announce the formation of our specialist Fusion Team, composed of recent hires Dr Ross Allen and Dr Cyd Cowley. They bring extensive domain expertise from across the fusion ecosystem: plasma science, systems engineering and design, advanced control, and advanced materials/manufacturing. They combine this knowledge with strong capabilities in the use of probabilistic Machine Learning to unlock new frontiers in this transformative, rapidly-evolving sector.
The digiLab Team
Welcome to the first digiLab newsletter! Here are some highlights of what we've been up to (so far) in 2024, and what's exciting us in the world of data-driven engineering and infrastructure. If you know someone who would be interested in these areas of Machine Learning, share our signup link!
The digiLab Team
digiLab hosted a Machine Learning (ML) workshop at Tokamak Energy, focusing on the application of mission-critical ML to fusion energy development workflows. Attendees discovered how to use digiLab’s ML platform, twinLab, to combine their fusion expertise with the latest in probabilistic modelling and uncertainty quantification (UQ).
The digiLab Team
Explainablity and transparency are key ideals-straight from Plato's realm of the forms-that developers must hold themselves to in order to facilitate proper ethical and sociological inspection of AI models. At digiLab, this is our bread and butter. This is why we have been registered onto the Ethical AI Database (EAIDB).
Dr Andy Corbett
Congratulations and a very warm welcome to the three students who successfully applied to our internship scheme this year, seeing off over 1600 other applicants to the post.
The digiLab Team
At digiLab we took the decision to provide our employees with a four day working week from day one. As a leader in the deep tech and machine learning community, it was the right thing to do for our growing business and our employees.
The digiLab Team
Are you looking to get the most accurate information from your models, and deploy them with confidence? Uncertainty quantification (UQ) is a set of essential computational tools that can help make sure your models are both accurate and reliable. And while UQ can be a complex subject, it doesn't have to be overwhelming! In this blog post, we will break down how Uncertainty Quantification works by discussing the concept of propagation of uncertainty.
The digiLab Team
"All models are wrong, but some are useful" is a famous quote attributed to the British Statistician George Box. In this series, we explore the field of Uncertainty Quantification (aka UQ), a field that seeks to quantify how wrong a model might be. Uncertainty Quantification is an increasingly important field as engineering and science seek to rely more heavily on simulations and machine learning algorithms to make critical decisions.
Prof Tim Dodwell