UNB Research

Dr. Stijn De Baerdemacker is the Canada Research Chair in Quantum Chemistry at UNB, as well as an associate professor of chemistry and associate research director of UNB’s Research Institute for Data Science and Artificial Intelligence.

We spoke with him about his research and what he hopes to accomplish as chair, and as part of that conversation, we talked to him about how he applies machine learning to chemistry and what doors quantum computing is opening for research.

Here's how he explained the importance of quantum computing:

To understand the relevance of quantum computing, we need to look at something called the many-body problem. The many-body problem basically has to do with complexity.

It can help to use people at a party as an example.

Let’s say I have one person in a room; perhaps they’re just sitting, enjoying a cup of tea or coffee. If we add a second person, two particles, we probably have a conversation, right?

But then, if we add a third person, we get a bit of a dynamic situation. And if you have ten people, suddenly you have a group. You can play soccer; you can have a debate; you can have many different conversations. Then if you have a thousand people, suddenly you have a soccer game with an audience—you can have a wave going across the stadium.

All of those individuals, if you put them each in a room, maybe they’re just sitting quietly drinking their tea. But their collective behaviour is different.

This is what I'm generally interested in. As researchers, we know that if we add more and more particles, the complexity and the behavior of those systems changes. We know that becomes more and more difficult to predict the outcomes of these systems, even for classical particles, such as soccer players and afficionados.

If I throw ten people together, I'm not sure whether they are going to kick a ball, or they are going to start knitting, or maybe they fight … it really depends on the internal degrees of freedom, like the interactions between the particles, or how much energy you have given to this group.

Now, let’s reduce the size of the system in our example to the atomic scale, and add the laws of quantum mechanics. When we take this many-body problem to that level, we see yet another big jump in complexity.

At the quantum level, this complexity has to do with something similar to the many-worlds theory. This theory is the idea that whenever you make a decision, all of the possible outcomes of that decision exist at the same time, each in its own universe; and so there are infinite universes for every possible combination of outcomes.

With the many-body problem, we have to likewise consider every possible outcome and interaction between these particles, at the same time. It’s likeTom Waits sings: “Everything you can think of is true ...” – but all at the same time!

Richard Feynman, the Nobel Prize winning theoretical physicist, once said that in order to understand how these quantum many-body systems work, we need a quantum system, a quantum computer, to compute the properties.

The good thing about quantum theory is that we mathematically understand this completely. We just have issues grappling with what the consequences are for the reality that comes out of that math.

It turns out—and this has been confirmed by multiple experiments—that our physical reality cannot be completely defined or predicted using deterministic laws. That is, there is an aspect of chance associated with each potential scenario or outcome. Sometimes, what should be the straightforward anticipated outcome simply isn’t. We only know the probability of what will happen.

Interestingly, because they operate as a quantum system in our physical reality, a quantum computer is also subject to these laws of probability. Running the same computing problem twice may yield two different outcomes, even with a perfect device.

Our task as quantum researchers is to exploit these probabilities in a way that they may lead to answers that can be obtained significantly faster than what our current computers can ever hope to achieve.

Quantum computing can be especially useful for us when the answer is simple, but the solution is difficult. To borrow a literary example, the answer may simply be 42, but the process for getting to that answer is very difficult, as you probably had to consider a gazillion number of scenarios to settle upon that 42, and that’s what we’re grappling with right now:

These particular problems are difficult because we need to keep track of, and account for, so many different scenarios throughout the solution.

A quantum device allows us to take all those different worlds and all those different scenarios, hold them all at once and then recombine everything at the end to get to 42—without actually having to test every possible outcome of every possible scenario.

It can do this because of the unique nature of quantum mechanics, where something can be in many states at the same time.

In essence, the same thing that makes quantum mechanics very confusing is the precise thing that we can potentially use to solve more complex problems.

For our particular purposes in chemistry, our hope is that at the end of the day, we can identify ways to test combinations of molecules and see what happens. Essentially, I just want to know if there’s a sign of a reaction, some sort of energy change, because that’s going to tell us if there’s something interesting or not. This is basically a complex question with a simple yes or no answer.

Right now, with our current supercomputers, I would need to do a huge number of computations to figure this out, which could easily take weeks or months to do in full. But I don’t need those computations or the individual results; I just need the final yes or no that tells me if there’s a reaction or not. And so, the way that quantum computers can manage all of these different scenarios and give us the end answer should, we hope, be perfectly suited to this kind of work.

More information

Dr. Stijn De Baerdemacker [UNB profile | CRC profile]| Department of Chemistry | Faculty of Science (Fredericton)

Research at UNB | Graduate Studies at UNB | Postdoctoral fellowships