TBNQuantum
High-fidelity quantum gate learning via physics-informed neural networks
Publication: quant-ph updates on arXiv.org
Author: Luis Fabra, Cristian E. Susa, Alcides Montoya, Ariel Norambuena
Source published: 04/11/2026, 05:00:00 UTC
Briefing created: 04/11/2026, 05:55:43 UTC
Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 04/11/2026, 05:13:24 UTC.
Briefings
A physics-informed neural network (PINN) successfully learns control fields for quantum gates using the time-dependent Schrödinger equation. The study achieved an average gate infidelity of 8.49e-8 for Hadamard gates and values between 3.39e-7 and 4.96e-6 for controlled-NOT gates under various interactions. The learned controls were represented by 139 coefficients with a minimum count of 24, and square-pulse strategies preserved polynomial references at high resolutions. Results were validated against the Lindblad master equation for dephasing and amplitude damping.
Key points
- PINNs trained on the time-dependent Schrödinger equation achieved an average gate infidelity of 8.49e-8 for Hadamard gates.
- Controlled-NOT gate infidelity ranged from 3.39e-7 to 4.96e-6 depending on interaction type and training configuration.
Context
This research applies physics-informed neural networks to quantum gate learning, utilizing the time-dependent Schrödinger equation as the governing model. The findings connect multi-state gate synthesis, control-resource characterization, and compact square-pulse playback, offering a method to synthesize control fields that satisfy physical constraints while minimizing gate errors.
Limitations
Results are based on unitary simulations and theoretical validation against the Lindblad master equation. The study covers one- and two-qubit systems but does not report results for larger systems or specific experimental setups beyond the tested dipolar scenarios.
Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 04/11/2026, 05:55:43 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.
View archived briefing on IPFS
bafybeiguifgm22q5fsslwfjaxn2txzontp35iowwcb4inyxthihhevlzc4