Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
VQE and You
VQE and You: past the cutting edge of materials science
Molecules and materials involving highly correlated electrons are hard to simulate classically but pose no great difficulty for a quantum computer. The Variational Quantum Eigensolver (VQE) process is the probable basis for future pharmaceutical and materials research on quantum computers.
VQE uses a classical machine learning process to discover ground state energies of a chemical molecule, accelerated by using a quantum computer to simulate the molecule. The classical design space has barely been investigated but has a huge impact on the performance of the VQE process. A good description is available at the IBM devblog.
We hope that Quantum Collective Knowledge (QCK ) will enable you to build the VQE methods that will power future chemistry, by providing an easy and reproducible environment to try new ideas. With QCK, you can design and benchmark new classical optimizers and quantum state preparation circuits (also known as ansatze), evaluate how good your solutions are (TTS), and compare your results to those contributed by others from all over the world.
The Time-To-Solution evaluation metric (TTS) quantifies the reliability of the result against the number of calls to the (expensive!) quantum resource used to calculate it. TTS is proportional to the number of quantum calls, while also penalises for failing to converge to the true minimum or for repeating your experiment too few times to determine reliability. A high TTS score indicates a clear need for improvement, while methods that achieve very low TTS scores may turn out to be suitable for use even on molecules that it is not possible to simulate on today's classical computers.