Reproduce “Universal expressiveness of variational quantum classifiers and quantum kernels for support vector machines” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jonas Jäger, Roman V. Krems
Researchers from the University of British Columbia and Technische Universität Darmstadt established the universal expressive power of Variational Quantum Classifiers (VQC) and Quantum Kernel Support Vector Machines (QSVM). They demonstrated that these quantum machine learning models can solve classification problems constructed from the k-Forrelation problem, which is known to be PromiseBQP-complete.
Key insights
- The specific design of the quantum feature map is paramount for unlocking the full computational power of QML models to solve quantum-hard problems.
- From a computational complexity perspective, both VQC and QSVM exhibit equivalent universal expressive power for PromiseBQP-complete problems.
- Quantum classifiers can, in principle, tackle problems at the theoretical limit of what quantum computers can efficiently solve (BQP class).
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.