MARCH 30, 2023

BRAID’S TEAM

<aside> 💡 Welcome to the second article in our series of Tech Team Profiles! Here, we’ll introduce you to the team building Braid’s automated engineering design technology. Below is an interview with Giles Strong, a Researcher at Braid with a background in particle physics and machine learning.

</aside>

Give a 1-minute elevator pitch of your PhD research.

I conducted my PhD research at the Laboratory of Instrumentation and Experimental Particle Physics in Lisbon, Portugal. The PhD was funded by the Horizon 2020 European Network. There were ten other PhDs around Europe working together on similar projects. The other PhD students and I were collectively researching how we can use deep learning algorithms to better understand particle collisions. I spent four years designing and training neural networks to understand how they can take in Large Hadron Collider (LHC) data and predict whether it’s a Higgs boson event or background noise.

Caption: Giles visiting CERN while he was a PhD student.

The Higgs boson was discovered in 2012. However, there’s still a lot of work to be done to measure its parameters, some of which require particle collisions in which two Higgs bosons are created simultaneously. Because such collisions are an incredibly rare event, distinguishing them amongst all of the other background noise is challenging. You need to collect many data points to have a high enough signal-to-noise ratio to confidently claim that you have observed such collisions. We used a combination of simulators, including a simulator of our entire particle collision detector, to compare to the data we collected at the LHC.

Because we could simulate both the collisions and signals, we knew the true class of each datapoint. We could then train a deep neural network to separate the signal from the background noise. This could be used to help estimate the rate at which the pairs of Higgs bosons are created at the real LHC, allowing us to better measure their properties.

Obviously your PhD research is a very complex topic for most people. How do you explain your research to a lay person with a limited technical background?

Metaphors can be somewhat useful, but every approach has its own pros and cons. Some researchers may argue that your explanation is too simplified or that you didn’t explain some very specific aspect that matters to them, but isn't actually relevant to the general public. Others may say that the metaphor is too complex and fails to deliver the intended message. Ultimately, you have to think about what you want to achieve by using metaphor. Are you trying to transfer specific knowledge or just have the listener gain an appreciation at a very high level?

As part of the funding for the PhD, there was support for scientific outreach. Every two weeks, each researcher in the group would write a blog post. For my own posts, I was taking quite specific points and explaining them, though ultimately the audience was not lay people. Fellow PhD student Alex Held wrote a nice post on the use of metaphors for outreach. His description of how the Higgs mechanism provides mass to particles can be used to help provide background to the general public for our research. As he explains towards the end, however, going too deep into the analogy causes it to break down.

What originally inspired you to study physics?

When I was at school, I read a series by Phillip Pullman called “His Dark Materials.” They’re fantasy novels based on physics. Later, I discovered the book by Mary and John Gribbin explaining the science behind the trilogy. There was also a television series by Brian Greene called “The Elegant Universe,” which was about quantum theory and string theory. Both the books and the television show presented physics in such a way that was unlike anything I had seen in class.

Caption: Giles visiting the Large Hadron Collider in 2008.

When I was a child, I knew that I wanted to go to CERN and that I wanted to do particle physics. When I was in senior school, I actually visited CERN on the last day before they switched on the Large Hadron Collider. I was able to go underground to see the particle accelerator. It was awe-inspiring. I studied physics at university and eventually ended up doing experiments with CERN.

Do you have any advice for someone like yourself who is considering making the leap from academia to industry?

I was on the fence about whether or not to transition from academia to industry. The career progression in academia appears to be that you get a postdoc and eventually a professorship, which involves a lot of lecturing. What I really want to focus on is the research. Working at Braid has been an interesting and rewarding opportunity to do that.

That said, I had a few worries about moving to industry: job interviews, office culture, concerns about "selling out", et cetera. So far, though, I feel it's been all the aspects of academia I enjoyed with fewer of the downsides. As a Researcher at Braid, I still have freedom over how I choose to approach problems. For anyone considering moving from academia to industry, I would say that the change is not as great as it might seem.