Creative Destruction Lab, a technology program affiliated with the University of Toronto’s Rotman School of Management in Toronto, Canada hopes to nurture numerous quantum learning machine start-ups in only a few years. This still new form of hybrid computing combines the computational speed and power of quantum computers with machine learning, the technical term for AI like Siri or Alexa.
Currently, researchers are mostly focused using the emergent technology of quantum computers to help machine learning programs to solve problems quicker or to use typical machine learning builds to add stability and potency to quantum computers. However, the end goal of either direction is to use many AI programs based on quantum computers to comprehend or better datasets and results from larger quantum calculations. Unfortunately, this goal will not be achieved until quantum computers are fully built and operational. While Google has a plan to build a 49-qubit machine by the end of the year, the hundred or thousand qubit computer that researchers hope for is still years or work away.
Despite being far afield, Peter Wittek, a researcher from the Institute of Photonic Sciences in Spain and academic director of the Creative Destruction Lab, says that doesn’t stop scientists from theorizing or even experimenting in the realm of quantum machine learning, a young field full of promise.
“To build universal quantum computers… is a big engineering challenge,” says Wittek. “But it turns out for quantum machine learning you need something less.” In the same way that quantum cryptography and quantum random number generation have been developed without large sized quantum computers, he says, so too could the field of quantum machine learning.
Wittek, who wrote **Quantum Machine Learning: What Quantum Computing Means to Data Mining, says that after the HHL quantum algorithm, named after its developers Aram Harrow, Avinathan Hassidim, and Seth Lloyd, was created, the field really came into its own. The algorithm solves massive linear algebra equations with many undefined variables in less time than any current supercomputer is capable. A large part of machine learning involves similar high-dimension algebra at which HHL excels and many researchers have flocked to HHL for this reason.
Although, even with all its unique properties, says Wittek, there are many besides HHL with perhaps more potential for sooner application in fields such as medicine, transportation, and finance.
Still, he adds, a quantum system will be a definite challenge to GPU-based machine learning, even if big names like Google and IBM can build a usable quantum computer. Machine learning, as it used today, is impressive enough.
On the other hand, when it comes to generating random numbers, typical machine learning falls short, according to Wittek. Specifically utilized in financial applications, Monte Carlo algorithms need truly random numbers to work ideally but conventional computing can only manage pseudo-randomness. This is where quantum machine learning could shine as they are designed around randomness.
Another advantage to quantum machine learning systems, according to Nathan Weibe, a researcher with Microsoft’s Quantum Architectures and Computation Group, is the use of a qubit versus the traditional binary bit system.
“If you think about a quantum computer, how do you understand what’s going on inside one?” Wiebe says. “The vectors that describe it exist in an incomprehensibly large space. There’s no way you can go through, read off every single entry of those vectors and figure out if the machine is working properly.”
While HHL has been very popular in recent technical literature, Scott Aaronson, a professor of computer science at the University of Texas at Austin, says that it has more value as hype than in application any time soon. In a 2015 review, Aaronson debates whether a “Buyer Beware” tag should accompany all quantum machine learning promises.
“Almost all the quantum machine learning algorithms that have been published over the last decade are really frameworks for algorithms,” Aaronson says. “They’re algorithms that don’t start with the classical problem that you would like to be solved and the answer to that problem.”
However, skepticism has not dulled enthusiasm for the possibilities in a quantum machine learning system and near-term application. The number of applications for the Creative Destruction Lab’s start-up boot camp in Toronto has exceeded expectations with 38 applicants for the 40 spots. As the application period is open until July 24th, it’s obvious that cutting edge entrepreneurs remain undaunted by neither the critics or the challenges ahead.
“Incorporation must be done by November, so these will be real companies,” Wittek says. “And the hope is by next summer we’ll have companies raising money.”
**Quantum machine learning is an emerging interdisciplinary research area at the intersection of quantum physics and machine learning. One can distinguish four different ways of merging the two parent disciplines. Quantum machine learning algorithms can use the advantages of quantum computation in order to improve classical methods of machine learning, for example by developing efficient implementations of expensive classical algorithms on a quantum computer.
More News to Read
- Will President Trump Be a Climate Crusader?
- Get Ready for the Chevy Bolt – It’s Nearly Here!
- How Can We Count Black Holes? How Many are There?
- Scientists Hope to Combat Rare Genetic Diseases With This Radical Proposal
- Neural Networks are Now Being Used to Represent Quantum Systems