Regensburg 2022 – wissenschaftliches Programm
QI 12.5: Vortrag
Donnerstag, 8. September 2022, 16:00–16:15, H8
Estimating molecular forces and other energy gradients efficiently on a quantum computer — •Michael Streif2, Thomas O’Brien1, Nicholas C. Rubin1, Raffaele Santagati2, Yuan Su1, William J. Huggins1, Joshua J. Goings1, Nikolaj Moll2, Elica Kyoseva2, Matthias Degroote2, Christofer S. Tautermann3, Joonho Lee1,4, Dominic W. Berry5, Nathan Wiebe6,7, and Ryan Babbush1 — 1Google Research, USA — 2Quantum Lab, Boehringer Ingelheim, Germany — 3Boehringer Ingelheim Pharma GmbH & Co KG, Germany — 4Department of Chemistry, Columbia University, USA — 5Department of Physics and Astronomy, Macquarie University, Australia — 6Department of Computer Science, University of Toronto, Canada — 7Pacific Northwest National Laboratory, USA
The calculation of energy derivatives underpins many fundamental properties for molecular systems, such as dipole moments or molecular forces. Nevertheless, most methods for quantum chemistry on quantum computers have focused on electronic structure calculations, even though energy derivatives are fundamental for many practical applications. Here, I will introduce quantum algorithms for the calculation of energy derivatives on noisy intermediate scale (NISQ) and fault tolerant (FTQC) quantum computers, with substantially reduced cost compared to previous methods. Our results suggest that the calculation of molecular forces has a similar cost to estimating energies. However, since molecular dynamics (MD) simulations typically require millions of force calculations, current known methods for MD on quantum computers are impractical and new approaches need to be found.