The XPRIZE Quantum Applications competition challenges teams to demonstrate quantum advantage on a real-world problem — not a contrived benchmark, not a toy demonstration. Drug discovery, financial optimization, and materials science are explicit target domains. The DaedArch Quantum team (Team ID 14342) is competing with a focus on molecular ground-state energy computation via Variational Quantum Eigensolver (VQE) with Active Space Reduction.
Standard VQE requires a quantum register large enough to encode the entire molecular orbital space — intractable on current NISQ-era hardware for any non-trivial molecule. Our active-space reduction pipeline pre-computes the chemically relevant orbital subspace using classical methods (CASSCF / coupled-cluster), then maps only that reduced space onto the quantum register. The classical pre-computation runs on GPU; the quantum step runs on cloud-accessible QPUs (IBM, IonQ, Quantinuum, Rigetti).
Quantum jobs queue across multiple providers with provider-specific preemption behavior. We engineered a preemption-safe orchestration layer that resumes from intermediate variational state vectors rather than re-running from scratch when a job is preempted. This unlocks substantially more QPU-time per dollar of cloud budget — the difference between a small team being competitive or not.
The active-space reduction technique we use for molecular ground-state energy is structurally identical to the dimensionality-reduction primitive in the broader Universal Verification Framework (Paper 5). When a verification problem has a high-dimensional state space, the framework's RC1–RC4 admissibility conditions apply only to the chemically (or physically) relevant subspace; the reduction to that subspace is what makes the framework computable in the first place.
Feedback from visitors, translated into business terminology and listed below. Use the assistant in the corner to add a comment.