Building Your Dream House- In 20 Microseconds
Imagine if you were building a house, and there were many features that you wanted to include, but you had to remain within a set budget? How would you determine the optimal combination of features that would maximize your satisfaction, without going over your budget?
Well, you could manually try out every single combination, but that could take thousands of years.
What if there was a way to look through all possible combinations and identify the best one, and that too, within your lifespan (unless we find a way to increase our lifespan to over 1000 years)?
Introducing quantum annealing.
Quantum annealing is a niche subsection of quantum computing. While many applications of quantum computing remain theoretical, quantum annealing is very relevant, and can be implemented now (More on that later!)
Founded in 1999, D-Wave was the world’s first commercial supplier of quantum computers. They build quantum annealers, which solve optimization problems.
But, you’re probably wondering- How are quantum annealers going to help me build my dream house, and that too, within my lifetime?
You see, quantum annealers leverage the laws of quantum physics, which are fundamentally based on finding low-energy states.
Essentially, the quantum annealer identifies all potential combinations for the house and plots it on a topographical map with peaks and valleys. From there, the quantum annealer looks for the global minimum, which is the lowest point on the entire graph.
On a classical computer, the process of identifying the minimum point could take thousands, or even millions of years- depending on the complexity of the problem and the number of variables and constraints.
TL;DR: With quantum computers that use qubits, we can efficiently sift through the map and find the optimal point very quickly.
A Quick Quantum Recap:
Here’s a crash course on quantum computing if you need a (qu)bit of a review (Feel free to skip this video if you’re familiar with quantum computing!)
But… How Do Quantum Annealers Actually Work?
Individual qubits have their own magnetic fields that can be influenced by external factors. These qubits are initially in superposition.
Through the process of quantum annealing, qubits in superposition collapse into either 1 or 0, which creates the double-well potential.
In the diagram above, the probabilities of 0 and 1 are equivalent, but as mentioned before, they can be influenced.
For example, by adding an external magnetic field, called a bias, the qubit minimizes its own magnetic force, which tilts the double-well potential, and consequently alters the probability.
Although adding biases is very powerful, true control over the quantum system occurs when qubits are linked together using couplers.
Couplers exploit another quantum phenomenon, called entanglement. Two entangled qubits are connected and will always be in the same state. Change in one qubit will immediately signal a change in the other, even if there is an entire universe between the qubits.
Couplers link two qubits together, creating either a positive correlation (Both qubits in the same state- 00 or 11) or a negative correlation (Opposite states- 01 or 10).
The linked qubits make that set of possibilities energetically favourable. For example, if a coupler wants to make matching qubit states more favourable, on an energy level graph, it would lower those two probabilities. The same process would occur for making opposite qubit states favourable.
Biases and couplers can be customized with unique strengths and directions to fit the problem at hand. This creates large quantum systems, with each state having a unique probability. Based on the topographic energy graph created, the quantum annealer identifies the optimal minimum energy state.
This entire process takes about 20 microseconds.
That’s right- The dream house that maximizes your happiness AND the value of your bank account could be created in only 20 microseconds.
The variability in customizing strengths and directions of the biases and couplers are precisely what make quantum annealers so versatile; giving them the ability to solve a wide range of optimization problems across a myriad of industries.
Quantum Annealing in Traffic Optimization
Current GPS systems have started to take some traffic into account, but millions of vehicles and pedestrians cause major congestion, especially within urban cities. In 2016, Volkswagen started tackling the issue of traffic congestion.
Depending on the region, there could be a huge number of factors involved; from vehicles and passengers to routes and destinations. Developing an optimal plan for use on a classical computer is nearly impossible.
So, Volkswagen, in conjunction with D-Wave, used the D-Wave 2000Q™ quantum annealer, along with data from 418 taxi cabs from Beijing, to develop a system that optimizes traffic flow.
“What makes the now developed solution so special is the possibility to scale it to any city. Size, infrastructure, traffic volume — quantum computers enable us to adapt our solution to all conceivable conditions,” says Florian Neukart, the Principal Scientist at Volkswagen’s CODE Lab in San Francisco.
This proof of concept was put to use in Lisbon, during Web Summit in 2019. During these 3 days, the city had an additional 70, 000 visitors, creating the perfect testing grounds for the quantum system.
The traffic optimization system was based on a combination of annealing and machine learning processes and calculated the fastest routes in near real-time.
Although quantum computing has been vastly theoretical, this application paved the way for the many prospects of quantum annealing in the future. From streamlining supply chains to determining the most efficient wing shape for airplanes, quantum annealing has the ability to revolutionize logistical solutions in nearly every field.
It’s interesting how experts consider quantum annealing to be the ‘least powerful and most narrowly applied’ form of quantum computing.
If the weakest form of quantum computing has such impactful applications, imagine the impact of fully developed universal quantum computers.
Key Takeaways:
- Quantum annealers are a specific type of quantum computers that solve optimization problems
- The process begins with a large set of qubits in the state of superposition (Some combination of 1 and 0)
- Qubits undergo the process of quantum annealing, where biases and couplers are added, forming large quantum objects
- These objects consequently result in altered probabilities of individual states
- The annealer identifies the minimum(optimal) energy state
- Volkswagon partnered with Google and D-Wave to optimize traffic in Beijing
- Quantum annealers show numerous prospects for the future of logistics and optimization
