MENU

QodeX Quantum and Partners: Forecasting Unstructured PDEs on 52-Qubit IBM Hardware

Quantum Computing Report USA
Overview
A research collaboration has demonstrated gradient-free forecasting of nonlinear partial differential equations (PDEs) on unstructured engineering meshes using a 52-qubit IBM Quantum Heron r3 processor. The hybrid quantum reservoir network (QRN) reduced forecast error by 21% to 41% compared to classical latent linear baselines. This approach decouples complex boundary geometries from quantum circuit topology, enabling the simulation of fluid dynamics on non-uniform grids.
In Depth

Key Findings

The research team, comprising QodeX Quantum, NVIDIA, RTX BBN, the University of South Carolina, and the Naval Surface Warfare Center Panama City Division, achieved a test NRMSE of 0.257 in 10-step closed-loop autoregressive rollouts on the cylinder flow dataset. This performance significantly outperformed a classical persistence baseline (1.126) and a noiseless CUDA-Q simulation (0.707), suggesting that physical hardware noise acted as an effective regularizer.

Technical Details

The team utilized a hybrid quantum reservoir network (QRN) on the ibm_boston processor. To handle unstructured meshes, the architecture pairs a pretrained graph neural network autoencoder with a fixed, untrained quantum reservoir. The reservoir is divided into dedicated memory and readout registers. The process involves:

  • Encoding latent flow variables into the memory register via data-reuploading rotations.
  • Coupling memory to readout qubits using tunable partial-SWAP gates.
  • Implementing mid-circuit measurement and reset of the readout register to create a tunable fading memory.
  • Applying a post-hoc Pauli-weight truncation method to manage readout shot requirements.

The system executed circuits up to 24 times longer than median qubit coherence times and sustained continuous evolution for up to 28 times median T1 and 24 times median T2 without collapsing the memory register.

Background & Context

Traditional gate-based quantum solvers for computational fluid dynamics have been limited by requirements for uniform grids, low Reynolds numbers, or shallow variational circuits. These limitations often lead to heavy measurement overheads or “barren plateaus.” The QRN approach bypasses these issues by using a fixed reservoir rather than a variational circuit.

Significance & Outlook

This study validates the ability to forecast nonlinear PDEs on complex engineering meshes using physical quantum hardware. Additionally, QodeX Quantum used this research to develop and validate modules for AlphaPsi, its agentic quantum workflow platform. The implementation code has been hosted on GitHub and the technical manuscript is available on arXiv.

Source: https://quantumcomputingreport.com/qodex-quantum-nvidia-rtx-bbn-and-u-s-navy-researchers-forecast-unstructured-pdes-on-52-qubit-ibm-hardware/

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.

  • Your email and selected fields are used only to deliver the newsletter.
  • We never share your information with third parties.
  • You can unsubscribe anytime via the link in each email.

See our Privacy Policy for details.

Takes about a minute · Unsubscribe anytime

Published by Troy-Technical, an independent site run by one engineer with a career in materials development.
About the author / Contact info@troy-technical.jp
Let's share this post !

Author of this article

TOC