A new computing model developed at the Cleveland Clinic can process certain tasks up to 500 times faster than other methods, according to a report by Phys.org. The approach, called quantum hyperdimensional computing, or QHDC, draws on principles from both neuroscience and quantum physics to build a framework designed to work naturally on quantum hardware rather than borrowing ideas from classical computers.
The work was led by Fabio Cumbo, a research associate in the lab of Daniel Blankenberg at Cleveland Clinic's Computational Life Sciences division. Cumbo published the first-ever implementation of QHDC in npj Unconventional Computing, covering two distinct experiments.
The foundation of the model comes from hyperdimensional computing, a type of computing rooted in neuroscience. The core idea is that the brain does not store information in a single neuron. When a person thinks of a cat, that concept is distributed across thousands or millions of neurons. If one neuron fails, the memory survives. In the same way, HDC spreads data across long vectors containing thousands of dimensions, so the system can still produce accurate results even when one vector contains an error.
QHDC takes that framework and runs it on quantum hardware. Quantum computers draw their power from qubits, which use the principle of superposition to exist in multiple states at the same time. By combining HDC's high-dimensional representations with quantum superposition, QHDC can encode and process complex data spaces far more efficiently than either approach alone.
"Most quantum computing software is still built by borrowing ideas from classical computing," Cumbo said. "I had the idea to explore a type of computation that works naturally on a quantum computer, rather than forcing it to fit a classical framework."
Blankenberg, the senior author of the paper, noted that even as quantum computing grows rapidly, researchers are still working out how to build frameworks and algorithms that take full advantage of what quantum hardware can actually do. Current quantum versions of artificial intelligence and neural networks involve complicated workflows that can take a long time to develop and run.
The team tested the QHDC framework across three environments: a classical computer, an idealized quantum simulator, and an actual quantum computer. That three-way comparison allowed researchers to evaluate not just raw speed but also how the framework performed under real-world quantum conditions, which are noisier and less predictable than simulations.
The researchers identified biomedical research as a particularly strong use case for QHDC. That field generates complex data with outcomes that are often difficult or impossible to predict in advance, which is exactly the kind of problem that high-dimensional quantum processing is suited to handle. Cumbo said that previous work he had done on hyperdimensional computing helped him recognize how naturally it mapped onto quantum systems, which eventually led to the QHDC concept.
The publication in npj Unconventional Computing marks the first formal implementation of the model and sets a foundation for future experiments testing QHDC on larger and more complex biomedical datasets.
