Posted by
(opens in new tab)For years now, Microsoft researchers have been working with academics and scientists to unlock the riddles of quantum computing (opens in new tab), a field that aims to merge the mysterious properties of quantum mechanics with computing. If achieved, a scalable quantum computer could rapidly accelerate how information is analyzed and processed, creating new forms of economic value.
Indeed, some have ventured that the move from classical computing to quantum computing could be as revolutionary as the shift from vacuum tubes to silicon transistors.
Ideas: Exploring AI frontiers with Rafah Hosn
Energized by disruption, partner group product manager Rafah Hosn is helping to drive scientific advancement in AI for Microsoft. She talks about the mindset needed to work at the frontiers of AI and how the research-to-product pipeline is changing in the GenAI era.
Given such stakes, it’s no wonder that Microsoft researchers are working feverishly to explore the mysteries the field holds, and one of those busy researchers is Krysta Svore (opens in new tab) (@krystasvore (opens in new tab)), the subject of the latest in the Microsoft Research Luminaries (opens in new tab) video series presented by Channel 9 (opens in new tab).
You’ll no doubt get a taste of the excitement by watching the complete video, but even a brief slice of it provides insights into Microsoft’s approach to quantum computing, which includes the use of topological qubits, and why Svore and her team are developing LIQUi|> (opens in new tab), a full software architecture for quantum computing.
In this video snippet, Svore discusses a potential future for machine learning:
“Quantum computing for machine learning is something we’re actively looking at right now,” she says. “We know quantum computers give exponential speedups for some problems. The question is: Can we get exponential speedups for problems in big data, data analysis, and machine learning? We have some new results showing that we can get speedups, quadratic speedups, with various classification problems.”
In this instance, though, she explains, speed might not be the ultimate benefit that quantum computing could deliver.
“In terms of machine learning and quantum computing,” Svore says, “what questions can we ask when we have a quantum computer that are different from the questions we ask traditionally? We could have new probability distributions, maybe new inference techniques. What does that give us for machine learning? It could give us much better models of users, systems … whatever we’re studying.
“What I think is really exciting in the area of machine learning and quantum computing is the ability to ask these new questions, not necessarily just look for these speedups. Speedups might not be as important as being able to ask a new question. This could really mean a huge shift in what we can do with machine learning. That’s an area we’re really excited about.”