AI and machine learning take a quantum leap

An epicentre of activity

Originally from Canada, Ferrie’s interest in quantum computing was first piqued as a teenager while flipping through an issue of Popular Science. He came across a story about this new type of computer, one he’d never heard of before. 

“I remember reading about how the computing happens in parallel universes and thinking, ‘wow, that’s really cool’,” he says. 

He nurtured this interest during his undergraduate degree and seized opportunities to combine his interests in information systems and quantum theory through postgraduate study at the University of Waterloo, where he obtained his PhD in quantum information.

Australia’s booming quantum community caught his eye. This was “the place to be if you were a scientist interested in quantum computing”, according to Ferrie. He’s worked here as a researcher and professor for the past 10 years within the UTS School of Computer Science in the Faculty of Engineering and IT. 

“Australians love to say they’re punching above their weight, and quantum is definitely an area where that’s true,” he says.

This isn’t just anecdotal; the Australian Government has made this emerging field a priority area, over the past several years, and its first State of Australian Quantum Report, released in 2024, reveals a deep commitment to growing this sector through private and public investment. 

“There’s lots happening in the whole Sydney ecosystem; Queensland and Victoria are also big hubs,” Ferrie says.

Learning new languages

Ferrie’s research focuses on quantum information theory, and in this field there are three main components: quantum artificial intelligence (AI), quantum machine learning and quantum control. These three combine into the systems that will, eventually, get quantum computers to do the things we want them to do. 

At the moment, quantum computers are “a bit of a black box”, he says. They have inputs, they have outputs, but everything that happens in between is still a bit of a mystery. 

“We’re trying to build these devices and systems from scratch, so we need to understand what it is we’re dealing with, how to control it, and then how to verify that it did what we wanted it to do,” Ferrie says.

The challenges to creating a fully functioning quantum computer, even a small one, are many. There’s building the actual physical device and all its supporting infrastructure, which is mostly the domain of electronics engineers and materials scientists. But Ferrie is concerned with the layered, more theoretical challenges of creating software to run on a machine that doesn’t yet exist. 

“We can come up with ideas for how the software works, the architecture, how we would verify it. It’s much more theoretical. We can’t create a program and run it on a device to see how it works,” he says.

Complicating this is uncertainty about how quantum computers of the future will be constructed given differences in opinion about materials, as well as best use cases how they might be integrated into existing technology. 

“When we use our computers today, we’re using high-level software that has many, many layers before you get down to the transistors that do the actual computations,” he says. 

“But with quantum systems, we don’t have those higher-level quantum programming languages and software. So right now, we’re working on the very low-level systems, which are informed by what the computer will be made of – and there are currently a lot of options for that.”

Taking a more targeted approach

To this end, he and his peers at QSI are working on projects that test various capabilities and applications of quantum systems. 

One project involves encoding ChatGPT into a quantum system to test its limits and better understand the resources required to train it. Another is looking at applying machine learning to controlling and characterising small-scale quantum prototypes. 

There are several smaller working groups within the centre tackling various niche applications as well. One is investigating quantum simulations to replicate chemical reactions, which has use cases in fields such as pharmaceuticals. Another is focused on creating software architectures for quantum computers to address future challenges like error correction. 

“We come up with the theoretical models, the protocols, the procedures, and the ideal situation is that we make it clear enough so that any experimental group can apply it or modify it to work on their real device,” Ferrie says. 

Bringing quantum computing to the classroom

Ferrie’s research is all part of a bigger quantum computing narrative that has captured the imaginations of the public, so much so that it’s often referenced as the ‘space race’ of today’s day and age. 

Part of getting the investment and interest required to really scale up development is engaging future engineers, scientists and the public with these concepts. As an academic, Ferrie is always looking for ways to better teach quantum software concepts to his students.

“At UTS, we’re unique in the sense that our quantum group doesn’t sit within science or physics – we sit within computer science,” he says.  

Ferrie takes a very practical approach using methods that wouldn’t be foreign to someone studying, say, software engineering for digital computers. 

But he ran into a roadblock: with digital computers, you can build a program or algorithm and test it on a device. We don’t yet have the capability to test and learn in this way for quantum software, at least not without needing to dive deep into the physics of the system. 

Ferrie wanted to keep things simpler, and as the saying goes: necessity breeds innovation. He and a colleague at QSI created an emulator so quantum software students could test their code without fear of missing something. 

“We wanted it to emulate not the existing, imperfect devices, but a future-state, perfect quantum computer so that the programs our students were creating actually run as designed,” he says. 

“The students enjoy it. Having something tangible in front of you that you can interact with, it changes the way you learn.”

It’s a tiny device, about the size of a hockey puck, and it allows a user to build and test software or even just play around with quantum concepts. It’s meant for fun and experimentation, and the team chose the name ‘Quokka’ to honour its Australian heritage. 

After seeing the response from students, the team saw the potential of this device to engage high school and grade school students with quantum concepts. The device comes loaded with educational materials and resources for teachers and students to dive deeper into the realm of quantum computing. 

The team has since commercialised the device under a company called Eigensystems. In the future, Ferrie is focused on getting this first product off the ground, but says they would like to develop more powerful products for research applications in the future. 

Education for the masses

Seeing a Quokka device in every classroom is just a small part of Ferrie’s mission to democratise quantum computing for the masses. There’s a lot of misinformation out there about what quantum physics is, what quantum computers can and can’t do, and it’s easy for investors, funding agencies and the public to be misled. 

“I think there are broader misconceptions about quantum computing and that’s how you end up with things like quantum healing crystals,” he says. 

He’s authored several books for adults to call BS on some of the field’s biggest misconceptions, as well as explain the realities of how quantum computing will intersect with our lives. 

But the titles he’s best known for are aimed at the littlest learners. Ferrie has authored an entire series of children’s books that explain quantum physics and other scientific concepts. Think titles like Quantum Physics for Babies, Neural Networks for Babies and Quantum Information for Babies. 

All up, he has more than 60 books in his oeuvre. The idea stemmed from wanting to write books for his own kids that explained what he spent his days doing. He focused on physics to start with, but the series became so popular it branched out into chemistry, engineering, biology and general STEM topics. 

“I wanted to create something for the youngest audience possible – I just love the juxtaposition of the most complex concepts written in a simple way,” he says. 
 
“It’s about removing the intimidation barrier. Introducing it at a younger age lets people see it doesn’t have to be complicated. We need to integrate it into everyday life and talk about how these ideas actually underpin everything around us.”

An optimistic outlook

There’s no agreed-upon timeline for when the first quantum computer will come online, but the momentum is there, bright minds are pushing forward, and now it’s a matter of seeing how the technology will integrate into our lives.  

It’s something only time can tell. Thomas Watson, the former CEO of IBM, famously said: “There will never be a market in the world for more than five computers.” 

And in a way, he was right, says Ferrie.  

“This was back in the day when computers were the size of server room,” he says. 

“So he was right in the sense that there was never a market for more than five of those computers.

“The same will be true for quantum computing. It would be pretty naïve to say that that’s the limit of this technology, that it will always be something that is expensive and requires a huge amount of infrastructure to utilise.” 

When the day does come, it will likely feel like the introduction of the first smartphone, Ferrie says, and you’ll probably access its capabilities the same way you interact with cloud computing now. From there, we’ll probably see incremental improvements in the technology as storage space and processing power increase. 

“We had no idea what we were doing when we developed digital technology, but we progressed, and it’s very impressive. Quantum computers have the benefit of seeing that trajectory and knowing what works in developing this technology,” he says.

“There’s room for optimism that it could happen faster, but we are at the beginning stage still. Several decades from now, though, we’ll have something that just looks miraculous from today’s standards.”