By Kevin Jackson for Quantinuum

The world is a lot smaller than it was in the previous century – or even in the previous decade. 

Customers are now accustomed to a wide variety of products that can be delivered from distributors all over the globe. While this is a great opportunity for suppliers, it also presents a challenge in the form of supply chain, logistics, routing, and optimization. 

How can distribution companies continue to serve the needs of their customers in the most efficient and effective way possible? This may seem like a simple question, but it becomes a complex computational problem when trying to account for all the variables that can occur within a distribution network. 

What’s more, classical computers simply cannot adequately perform this optimization calculation in real-world scenarios. Because of the number of variables, the math just runs too slow. 

That said, new work in quantum computing has shown promise in applications within the optimization field. To that end, we interviewed Quantinuum’s Megan Kohagen and Dr. Mattia Fiorentini to better understand how quantum computing could to optimized logistics and supply chains.

Kohagen and Fiorentini are participating in a panel about quantum computing at Manifest: The Future of Logistics conference this week in Las Vegas, Nevada.

Beyond classical computing

When it comes to optimization it is all about maximizing or minimizing an objective.  A good example is a company that delivers goods and products but owns a limited number of trucks. To improve efficiency and minimize costs, the company needs to maximize the number of objects its trucks carry and identify the shortest routes between deliveries.

“You have all these constraints, you have your objective, and you’ve got to make decisions,” said Kohagen, an optimization researcher. “The decisions end up being things like how many goods you are going to send between your distribution centers and your stores? Each of these optimization problems, even if you consider them separately, are hard problems. The technical term is that they’re (non-deterministic polynomial)-hard because you’re dealing with discrete things. For example, I can’t send half a T-shirt to my customer. I can only operate with whole integers.” 

Fiorentini expands on this: “In logistics, we cannot leave anyone behind. If we need to deliver medicine, we cannot decide ‘the villages with less than 1,000 people – we don’t supply them. There are too many, and not enough people live there’. That’s not an option in today’s world.”

Today’s computers struggle to solve these NP-hard optimization problems because of the number of ever-changing variables.  Consider the much-studied Traveling Salesperson Problem, which is often used to illustrate the complexity of managing logistics, routing, and supply chains.  

This is a theoretical problem where a machine is tasked with finding the shortest route between an identified list of cities that a “salesperson” must visit before returning to the point of origin. This problem is simple enough with only a few cities, but it becomes exponentially harder as more locations are added, and other factors such as multiple salespeople, weather conditions, and unforeseen events arise. 

Classical computers can solve this theoretical problem for a single salesperson traveling to thousands of cities. But this scenario is not realistic, and this is where classical computers begin to struggle.

“The Traveling Salesperson Problem is not very representative of what happens in the real world,” Kohagen said. “For example, with online ordering so prevalent, a retailer has orders coming in constantly. They must determine how to efficiently retrieve those items from the warehouse, pack them into the trucks, and then transport them to the customers.”

Today, the reality of an extended supply chain or distribution network is beyond what the best classical computer can solve. Quantum computers harness unique properties of quantum physics that enable them to examine all possible answers simultaneously and then concentrate the probable output of the computation onto the best option.

“Classical is a great technology, but it doesn’t cut it here,” said Fiorentini, who develops and tests quantum algorithms for optimization. “Quantum is the best alternative to classical computing that we have.“

The quantum computing opportunity

Optimization problems have long been viewed as “killer applications” for quantum computing and research conducted by Fiorentini, Kohagen and others has begun to prove that. 

Fiorentini believes it is time for decision makers to explore and invest in quantum-enabled solutions for optimization problems. “There are two decisions here for decision makers,” he said. “We either give up on the problem and say, ‘we’ll just do the best we can with a classical solution, or we start allocating a budget for really developing quantum technology.”

Quantum computing is expanding rapidly and is poised to disrupt markets such as optimization.  A similar situation is the power sector, which is experiencing major disruptions due to innovations in renewable energy resources, energy storage, and regulatory reform. 

Every technology has a tipping point, and all signs point to a current trend in quantum computing moving rapidly to real-world applications in optimization.

“There are a lot of algorithms being developed for optimization right now,” said Kohagen. “If you really want to advance your business with quantum methods for logistics or supply chain, this is the moment to start. Decision makers must act quickly. Those that seize the opportunity before others will have a major advantage over those who lag.”

“As quantum computers continue to scale in computational power, they’ll be able to handle increasingly complex calculations to deliver more robust and optimized supply chain solutions,” said Tony Uttley, President and COO of Quantinuum.

“We’re excited by the acceleration of our System Model H1 technologies, Powered by Honeywell. Measured in terms of qubit number as well as quantum volume, we’re meeting our commitment to increase performance by a factor of 10X each year,” he said. “Alongside other revolutionary advances such as real-time error correction, we look forward to supporting the commercialization of quantum applications that will change the way logistical challenges are met. In fact, within the coming few months we’ll be sharing more exciting news regarding our latest technological achievements.”

Want to learn about our work to develop quantum-enabled optimization solutions for companies? Contact our experts

Since 1851, the World’s Fairs, now known as World Expos, have brought people together to innovate, collaborate and solve important problems. 

The theme for this year’s World Expo, currently taking place in Dubai in the United Arab Emirates, is “Connecting Minds, Creating the Future.” Quantinuum’s trapped-ion quantum computing technology, Powered by Honeywell, is one of a handful of breakthrough technologies chosen to showcase American innovation at the global event. 

Honeywell’s quantum computing division recently combined with Cambridge Quantum to form Quantinuum. Honeywell is majority owner of the new company, which aims to accelerate the development of this disruptive technology and deliver real-world, quantum solutions faster.

Here are some things to know about the World Expo, quantum computing and our involvement in the global event:

1. This is the first World Expo to showcase quantum computing. Visitors to the USA Pavilion take a 25-minute journey along a moving walkway and experience an immersive exhibit about how American ingenuity is shaping the future. Themed “The Sky is No Longer the Limit,” the exhibit showcases the System Model H1 quantum computing technology and its potential to help humans solve problems considered too complex for classical computers.
   ‍
2. The United Arab Emirates is the first Middle Eastern country to host the World Expo. The first World’s Fair took place in London in 1851. It is held every five years and much like the Olympics, countries vie to host the event. In 2013, the United Arab Emirates became the first country in the Middle East, Africa, and South Asia (MEASA) region) selected to host a World Expo.
   ‍
3. The World Expo site in Dubai covers more than 1,000 acres. The USA Pavilion boasts more than 20,000 square feet of indoor and outdoor exhibition space. 
   ‍
4. There are expected to be 25 million visits to the Expo during its six-month run. More than 4 million are expected to visit the USA Pavilion.
   ‍
5. Honeywell is an official sponsor of the USA Pavilion, which is also equipped with some of the company’s advanced security and access control systems.
   ‍

“We are proud to showcase quantum computing technology at what has been the world’s premier event for innovation for more than 170 years,” said Tony Uttley, president of Quantinuum. “Quantum computing will disrupt our world and enable us to solve critical challenges that we can’t with today’s technology. It is an honor that our trapped-ion quantum computing technology, which has led to our release of the highest performing commercial quantum computers in the world, has been included in momentous event.” 

Quantinuum’s H-Series quantum computers, Powered by Honeywell, continue to deliver on exponential performance gains

Over the course of 2021, Quantinuum’s customers and collaborators were the beneficiaries of a deliberate, strategic approach to quantum computing design. Namely, that it is possible to release a generation of quantum computers that can be quickly and systematically upgraded in parallel with commercial usage, allowing customers immediate access to the latest upgrades.

With the release of the System Model H1, Powered by Honeywell, in fall 2020, Quantinuum began a real-time demonstration of its design approach. The first System Model H1, referred to as the H1-1, launched in October 2020 with a measured quantum volume of 128. Quantum volume is a metric introduced by IBM to measure the overall capability and performance of a quantum computing system regardless of technology. (Calculating quantum volume requires running a series of complex random circuits and performing a statistical test on the results.) 

During 2021 Quantinuum, under its trapped-ion hardware group, previously known as Honeywell Quantum Solutions, made multiple upgrades to the H1-1 achieving measured quantum volume records of 512 in March 2021 and the 1,024 in July 2021. During that same period, Quantinuum was quietly releasing its second H1 generation quantum computer to customers and collaborators, called the H1-2. The System Model H1-2 uses the same ion-trap architecture, control system design, integrated optics, and photonics as the H1-1. 

“Our H1 generation of quantum computers are nearly identical copies, with the ongoing exception that at any given time one computer might have received upgrades prior to the other,” said Dr. Russ Stutz, Head of Commercial Products for the hardware team. “Our goal is to provide users with the highest performing hardware as they work on solving real world problems."

Upgrades to both H1 quantum computers over the course of 2021 included improved gate and measurement fidelities, reduced memory errors, faster circuit compilation, inclusion of real-time classical computing resources and quantum operations using 12 qubits, versus the 10 qubits available at initial release.

What has been remarkable about the approach, is the ability to deliver near-continuous capability upgrades while being consistent on performance. 

“Our customers frequently comment about their ability to reliably get expected results, including when running deep circuits and using sophisticated features like mid-circuit measurement, qubit reuse and conditional logic,” said Dr. Brian Neyenhuis, Head of Commercial Operations for the hardware team.

Just this past week, H1-2 measured a Quantum Volume of 2,048 (211), setting a new bar on the highest quantum volume ever measured on a quantum computer. The performance of the H1 generation of quantum computers continues to achieve the 10X per year increase that was announced in March 2020.

The Data

The average single-qubit gate fidelity for this milestone was 99.996(2)%, the average two-qubit gate fidelity was 99.77(9)%, and state preparation and measurement (SPAM) fidelity was 99.61(2)%. We ran 2,000 randomly generated quantum volume circuits with 5 shots each, using standard optimization techniques to yield an average of 122 two-qubit gates per circuit.

The System Model H1-2 successfully passed the quantum volume 2,048 benchmark, returning heavy outputs 69.76% of the time, which is above the 2/3 threshold with 99.87% confidence.

The plot above shows the heavy outputs for Quantinuum’s tests of quantum volume and the dates when each test passed. All tests are above the 2/3 threshold to pass the respective quantum volume benchmark. Circles indicate heavy output averages and the violin plots show the histogram distributions. Data colored in blue show system performance results and red points correspond to modeled, noise-included simulation data. White markers are the lower two-sigma error bounds.

The plot above shows the individual heavy outputs for each quantum volume 2,048 circuit. The blue line is an average of heavy outputs and the orange line is the lower two-sigma error bar which crosses the 2/3 threshold after 818 circuits, which corresponds to passing.

This is the latest in a string of accomplishments for Quantinuum, which recently announced the completion of its combination between Honeywell Quantum Solutions and Cambridge Quantum Computing to form the largest stand-alone integrated quantum computing company in the world. This news also falls on the heels of the release of Quantinuum’s flagship product, Quantum Origin, the world’s first quantum-enhanced cryptographic key generation platform. 

“We look forward to continued momentum in 2022 with expected advances in multiple application areas as well as further advances in the H-Series quantum computers”, said Tony Uttley, President and Chief Operating Officer of Quantinuum.

* The Honeywell trademark is used under license from Honeywell International Inc. Honeywell makes no representations or warranties with respect to this product or service.

Collaboration is at the core of any important technological development. From the steam engine to the internet, humanity’s innovations interweave themselves between seemingly disparate communities. 

That said, new technologies don’t always work together. There are many who still remember how Mac floppy disks were incompatible with PC machines, and vice versa. 

Quantum computing is no different, which is why Quantinuum is a founding member of the new Quantum Intermediate Representation (QIR) Alliance announced today by the Linux Foundation. The QIR alliance is working hard to ensure this technology reaches its full potential.

The siloed nature of early quantum computing developments has protected vital intellectual property, but it has also created a separation of resources. Quantum software from one organization may not work on the hardware of another, which can be an enormous obstacle for researchers. 

The QIR Alliance is solving this problem by establishing an intermediate representation to enable interoperability within the quantum ecosystem. Based on the open source LLVM intermediate language, the QIR Alliance will create a standard set of rules for representing quantum constructs consistent with LLVM data model. 

In doing so, the QIR Alliance hopes to enable wider collaboration and a quantum community built around principals of interoperability. 

How does intermediate representation (IR) work? 

Although programming languages may look like machine speak to the untrained eye, these languages are for the human programmers. Intermediate representation approach splits the compilation process into two parts. A user language compiler converts human-readable program representation into IR. A hardware-specific compiler takes the IR and converts it into a set of machine-level instructions that the computer can understand. 

This approach allows a hardware-specific compiler to work with many different source languages and still give the machine adequate instructions that it can comprehend. Conversely, quantum programming language developers only need to compile their new languages to one IR representation to run on many different machines. This enables innovation on both sides of the ecosystem while avoiding duplication of effort.

Therefore, a compiler-level solution makes sense to achieve the collaborative goals the QIR Alliance has set out. 

LLVM is a collection of compiler and toolchain technologies that are designed around a language-independent intermediate representation. This common platform allows many source languages to share optimizers and executable generators, which enables a large amount of re-use in compiler machinery. 

In short, this should allow quantum hardware to work with more varieties of software than they previously could. Rather than having to rewrite software based on the specific machine researchers want to use, the QIR Alliance will allow much more collaboration from previously disparate organizations. 

An additional interesting part of LLVM is that it also facilitates integration with many languages and tools built for classical computation environments. While quantum and classical computers may seem like competing technologies, many researchers expect to see quantum and classical computing resources working together in the future. The use of LLVM will facilitate quantum and classical computations interaction at the hardware level. 

What’s the benefit? 

For an organization like Quantinuum, the QIR Alliance offers several enticing advantages. 

To begin, this initiative will benefit the current quantum ecosystem. As the reality of quantum machines begins to truly materialize, it is no longer feasible for researchers to work with systems that are not interoperable. Much like how Mac floppy disks were once not compatible with PC machines, the quantum industry will need to come together to create a valuable product for the consumer. 

On top of this, the quantum sector must be constantly looking to the future and how this technology could improve and change in the coming years. All the major players within the quantum ecosystem must adopt a forward-thinking approach to intermediate representation that will fulfill the needs of current machines while also staying mindful of yet-to-be-developed hardware. 

Keeping an eye on the horizon is a goal of the QIR Alliance, and Quantinuum is fortunate to be a part of such an important step in quantum computing’s history. 

• Quantum Origin is the world’s first commercial product built using quantum computers that delivers an outcome that classical computers could not achieve
• Quantum Origin is the first platform to derive cryptographic keys using the output of a quantum computer to ensure data is protected at foundational level against evolving attacks
• It provides immediate protection to enterprises and governments from current security issues, arising from the use of weaker random number generators (RNGs)
• Quantum Origin also helps protect against ‘hack now, decrypt later’ attacks, which are already happening and will have future implications
• The quantum-enhanced cryptographic keys generated by Quantum Origin are based on verifiable quantum randomness and can be integrated into existing systems. The protocol relies on “entanglement”, a unique feature of quantum mechanics.
• Quantum Origin supports traditional algorithms, such as RSA or AES, as well as post-quantum cryptography algorithms currently being standardized by the National Institute for Standards and Technology (NIST)

Cambridge Quantum (CQ), the global leader in quantum software, and a wholly owned subsidiary of Quantinuum, the world’s leading integrated quantum computing company, is pleased to announce that it is launching Quantum Origin – the world’s first commercially available cryptographic key generation platform based on verifiable quantum randomness. It is the first commercial product built using a noisy, intermediate-scale quantum (NISQ) computer and has been built to secure the world’s data from both current and advancing threats to current encryption.

Randomness is critical to securing current security solutions as well as protecting systems from the future threat of quantum attacks. These attacks will further weaken deterministic methods of random number generation, as well as methods that are not verifiably random and from a quantum source.

Today’s systems are protected by encryption standards such as RSA and AES. Their resilience is based on the inability to “break” a long string from a random number generator (RNG). Today’s RNGs, however, lack true, verifiable randomness; the numbers being generated aren’t as unpredictable as thought, and, as a result, such RNGs have been the point of failure in a growing number of cyber attacks. To add to this, the potential threat of quantum attacks is now raising the stakes further, incentivizing criminals to steal encrypted data passing over the internet, with a view to decrypting it later using quantum computers. So-called “hack now, decrypt later” attacks.

Quantum Origin is a cloud-hosted platform that protects against these current and future threats. It uses the unpredictable nature of quantum mechanics to generate cryptographic keys seeded with verifiable quantum randomness from Quantinuum’s H-Series quantum computers, Powered by Honeywell. It supports traditional algorithms, such as RSA or AES, as well as post-quantum cryptography algorithms currently being standardized by the National Institute for Standards and Technology (NIST).

“We have been working for a number of years now on a method to efficiently and effectively use the unique features of quantum computers in order to provide our customers with a defense against adversaries and criminals now and in the future once quantum computers are prevalent,” said Ilyas Khan, CEO of Quantinuum and Founder of Cambridge Quantum. He added “Quantum Origin gives us the ability to be safe from the most sophisticated and powerful threats today as well threats from quantum computers in the future.”

Duncan Jones, head of cybersecurity at Cambridge Quantum, said: “When we talk about protecting systems using quantum-powered technologies, we’re not just talking about protecting them from future threats. From large-scale takedowns of organizations, to nation state hackers and the worrying potential of ‘hack now, decrypt later’ attacks, the threats are very real today, and very much here to stay. Responsible enterprises need to deploy every defense possible to ensure maximum protection at the encryption level today and tomorrow.”

Quantum-enhanced keys on demand

With Quantum Origin, when an organization requires quantum-enhanced keys to be generated, it can now make a call via an API. Quantum Origin generates the keys before encrypting them with a transport key and securely relaying them back to the organization. To give organizations a high-level of assurance that their encryption keys are as unpredictable as possible, Quantum Origin tests the entire output from the quantum computers, ensuring that each key is seeded from verifiable quantum randomness.

These keys are then simple and easy to integrate within customers' existing systems because they’re provided in a format that can be consumed by traditional cybersecurity systems and hardware. This end-to-end approach ensures key generation is on-demand and is capable of scaling with use, all while remaining secure.

Quantum Origin in practice

Quantum Origin keys should be used in any scenario where there is a need for strong cybersecurity. At launch, Cambridge Quantum will offer Quantum Origin to financial services companies and vendors of cybersecurity products before expanding into other high priority sectors, such as telecommunications, energy, manufacturing, defense and government.

The technology has already been used in a series of projects with launch partners. Axiom Space used Quantum Origin to conduct a test of post-quantum encrypted communication between the ISS and Earth — sending the message “Hello Quantum World” back to earth encrypted with post-quantum keys seeded from verifiable quantum randomness. Fujitsu integrated Quantum Origin into its software-defined wide area network (SDWAN) using quantum-enhanced keys alongside traditional algorithms.

For more information:
‍
— Learn more about Quantum Origin
— Explore our partnerships
— Read about our case studies
— Book a demo with our experts

The new software development toolkit for quantum natural language processing tested and benchmarked on System Model H1 technology

Telling Alexa to play “Schrodinger’s Cat” by Tears for Fears. Asking Siri for directions to a quantum-themed bar or restaurant. A smart phone autocorrecting a word in a text message.

These are everyday applications of natural language processing – NLP for short – a field of artificial intelligence that focuses on training computers to understand words and conversations with the same reasoning as humans.

NLP technologies have advanced rapidly in recent years with the help of increasingly powerful computing clusters that can run language models that examine reams of text and count how often certain words appear. These models train devices to retrieve information, annotate text, translate words from one language to another, answer questions, and perform other tasks.

The next step is to “teach” computers to infer meaning, understand nuance, and grasp the context of conversations. To do that, however, requires massive computational resources and multiple algorithms or data structures.

A United Kingdom-based quantum computing company believes the answer lies with qubits, superposition, and entanglement.

Cambridge Quantum recently released lambeq, a new open-source software development toolkit, that enables researchers to convert sentences into quantum circuits that can be run on quantum computers. It is the first toolkit developed specifically for quantum natural language processing – or QNLP - and was tested on System Model H1 technology before it was released.

The software takes the text, parses it, and then uses linguistics and mathematics to differentiate between a verb, noun, preposition, adjectives, etc., and label them to understand the relationships between words.

Cambridge Quantum researchers tested 30 sentences on the System Model H1, which was able to classify words correctly 87 percent of the time.

“We deem that a success,” said Konstantinos Meichannetzidis, a member of the CQ team. “We found that our software works well with the Honeywell technology and were able to benchmark the performance of this quantum device.”

The lambeq project also represented a first for Honeywell Quantum Solutions. It was the first QNLP problem run on the System Model H1 hardware.

“We are really excited to be a part of this work and contribute to the development of this important toolkit,” said Tony Uttley, president of Honeywell Quantum Solutions. “Applications like this help us test our system and understand how well it performs solving different problems.”

(Honeywell Quantum Solutions and Cambridge Quantum have a long-standing history of partnering together on research and other projects that benefit end-customers. The two entities announced in June they are seeking regulatory approval to combine to form a new company.)

Why QNLP?

For humans, decoding conversations to understand meaning is a complex process. We infer meaning through tone of voice, body language, context, location, and other factors. For computers, which do not rely on heuristics, decoding language is even more complex.

The only way to create some sort of “meaning-aware” NLP is to explicitly encode compositional, semantic sentence structure into language models. To do this on a classical computer, however, requires massive computational resources, which are costly, and would likely still take months to process.

Quantum computers, on the other hand, run calculations and crunch data very differently.

They harness unique properties of quantum physics, specifically superposition and entanglement, to store and process information. Because of that, these systems can examine problems with multiple states and evaluate a large space of possible answers simultaneously.

What this means in terms of natural language processing is that quantum computers are likely to go beyond counting how often certain words appear or are used together. As noted above, quantum computers can identify words, label them as a noun, verb, preposition, etc., and understand the relationship between words. (lambeq uses the Distributional Compositional Categorical – or DisCoCat – model to do this.)

This enables the computer to infer meaning, and also provides insight into how and why the computer made connections between words. The latter is important for validating data and also expanding the use of QNLP in regulated sectors such as finance, legal, and medicine where transparency is critical.

Built upon previous work

The Cambridge Quantum team has long explored how quantum computing can advance natural language processing, and has published extensively on the topic.

In December 2020, researchers released two foundational papers that demonstrated that QNLP is inherently meaning-aware and can successfully interpret questions and respond.

Earlier this year, the team performed the first NLP experiment conducted on a quantum computer by converting more than 100 sentences into quantum circuits using an IBM technology. Researchers successfully trained two NLP models to classify words in sentences.

The release of lambeq and the testing of the open-source toolkit on the Honeywell System Model H1 represents the next steps in their QNLP efforts.

“Our team has been involved in foundational work that explores how quantum computers can be used to solve some of the most intractable problems in artificial intelligence,” said Bob Coecke, Cambridge Quantum’s chief scientist.

“In various papers published over the course of the past year,” Coecke added, “We have not only provided details on how quantum computers can enhance NLP but also demonstrated that QNLP is ‘quantum native,’ meaning the compositional structure governing language is mathematically the same as that governing quantum systems. This will ultimately move the world away from the current paradigm of AI that relies on brute force techniques that are opaque and approximate.”