Benchmark

Fermi-Hubbard VQE Optimizer Benchmark

VQE with Hamiltonian Variational Ansatz · Simulation · 12 qubits · Qiskit, PennyLane

Comprehensive benchmark of 30 classical optimizers on 372 instances of VQE for the Fermi-Hubbard system using the Hamiltonian variational ansatz. Best-performing optimizers are gradient-descent variants (Momentum, ADAM with finite differences), SPSA, CMA-ES, and BayesMGD. The study demonstrates that finite-difference step size has a very significant impact on VQE convergence quality.

Fermi-HubbardVQEoptimizer-benchmarkcondensed-mattersimulation

Comprehensive benchmark of 30 classical optimizers on 372 instances of VQE for the Fermi-Hubbard system using the Hamiltonian variational ansatz. Best-performing optimizers are gradient-descent variants (Momentum, ADAM with finite differences), SPSA, CMA-ES, and BayesMGD. The study demonstrates that finite-difference step size has a very significant impact on VQE convergence quality.

Key Metrics
Optimizers benchmarked
30
VQE instances tested
372
Best optimizers
Momentum, ADAM, SPSA, CMA-ES
Why It Matters

Largest systematic comparison of classical optimizers for VQE, revealing that finite-difference step size is the dominant factor in convergence quality.

Hardware

Simulator / hardware-agnostic

Framework

Qiskit, PennyLane