Cambridge Quantum Computing (CQC) announced on Tuesday its scientists have developed methods and demonstrated that quantum machines can learn to infer hidden data or information from broad probabilistic reasoning models. The new methods are designed to improve a wide variety of applications where complex systems are needed to account for uncertainty.
CQC researchers published a paper on the pre-print repository arXiV that established quantum computers can learn to deal with real world scenario-caused uncertainty.
Marcello Benedetti led the research team responsible for the paper. He co-authored it with Brian Coyle, Michael Lubasch, and Matthias Rosenkranz. These scientists work in the Quantum Machine Learning division of CQC, which is managed by Mattia Fiorentini.
The paper implemented three proofs of principal on an IBM Q quantum computer and on simulators in an effort to demonstrate quantum-assisted reasoning on inferences on random instance of a textbook Bayesian network, inferring market regime switches in a hidden Markov model of a simulated financial time series and a medical diagnosis task known as the “˜lung cancer“™ problem.
These proofs of principle imply quantum computers that use highly expressive inference models could enable new applications in multiple fields. For example, quantum machines could be used in medical diagnosis, fault-detection in mission-critical devices, or financial forecasting for investment management.
The paper builds on the fact that sampling from complex distributions is among the most promising means to achieve a quantum advantage in machine learning. This work illustrates how even early-stage quantum computing is an effective tool for studying science“™s most daunting questions, such as human reasoning emulation, according to CQC.
Quantum software and hardware developers, along with machine learning scientists, are expected to gain the most from the development in the short-term. Quantum devices are anticipated to improve in the coming years and CQC said the research has laid the foundation for quantum computing to be applied to probabilistic reasoning.