A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
The variational quantum eigensolver (VQE) is a hybrid quantum-classical algorithm that finds the ground-state energy of a Hamiltonian by variationally optimizing a parameterized quantum circuit (ansatz). The quantum computer prepares a trial state |psi(theta)> and measures the expectation value <psi(theta)|H|psi(theta)>; a classical optimizer updates theta to minimize this energy. By the variational principle, the measured energy is always an upper bound on the true ground-state energy. VQE is designed for near-term noisy intermediate-scale quantum (NISQ) devices, requiring shallow circuits that tolerate noise, at the cost of requiring many measurement repetitions and classical optimization iterations.
Quantum chemistry is one of the most promising near-term applications of quantum computing. The core computational problem is finding the ground-state energy of a molecular Hamiltonian — determining how electrons arrange themselves to minimize energy. For small molecules, classical computers handle this well, but the Hilbert space grows exponentially with the number of electrons, making exact classical solutions intractable beyond about 50 electrons. Quantum computers can represent quantum states natively, but fault-tolerant algorithms like quantum phase estimation require error-corrected qubits that are not yet available at scale. VQE fills this gap by running on noisy, near-term hardware.
The algorithm has a simple structure. Step 1: Choose a parameterized quantum circuit U(theta) — the ansatz — that maps |0>^n to a trial state |psi(theta)> = U(theta)|0>^n. Step 2: Prepare |psi(theta)> on the quantum computer and estimate E(theta) = <psi(theta)|H|psi(theta)> by decomposing H into a sum of Pauli strings and measuring each term's expectation value. Step 3: Feed E(theta) to a classical optimizer, which proposes new parameters theta'. Step 4: Repeat until convergence. The final E(theta*) is an upper bound on the ground-state energy, with equality when the ansatz can express the true ground state.
The decomposition of H into measurable terms is a key practical step. A molecular Hamiltonian in second-quantized form is mapped to qubits using transformations like Jordan-Wigner or Bravyi-Kitaev, producing a sum of O(N4) Pauli strings for N spin-orbitals. Each Pauli string's expectation value is estimated from repeated measurements (shots). The total number of measurements required can be very large — this measurement overhead is a significant bottleneck. Techniques like grouping commuting Pauli terms and using classical shadows reduce this cost.
VQE's main challenges are barren plateaus (exponentially vanishing gradients for deep or random circuits, making optimization intractable), noise (gate errors bias the energy estimate, though error mitigation techniques can partially compensate), and local minima in the optimization landscape. Despite these challenges, VQE has been demonstrated on real quantum hardware for small molecules (H2, LiH, BeH2) and remains a leading candidate for achieving quantum advantage in chemistry. The broader principle it embodies — using the quantum computer as a state preparation and measurement device while offloading optimization to a classical computer — is the template for all NISQ-era variational algorithms.