Extreme-Scale Ising Machines with Cluster Mean-Field Theory
Abstract
Scaling analog and digital Ising machines to larger problems requires overcoming finite device capacity and the cost of communication between devices. We present cluster mean-field theory (CMFT), a framework that partitions a large interaction graph into clusters sized to fit available hardware. Each cluster performs local updates independently, while interactions across cluster boundaries enter through periodically updated effective biases computed from boundary-spin averages. To reduce the error introduced by fixed cluster boundaries, we introduce dynamic partitioning, which cycles through multiple partitions so that interactions approximated by mean fields at one stage can act through instantaneous spins at another. On three-dimensional spin glasses and planted Pegasus instances, dynamic CMFT exhibits power-law decay of residual energy over sweep budgets. This shows that solution quality continues to improve with computational effort despite the mean-field approximation. An automated graph partitioner combined with a weighted recovery ratio provides a practical heuristic for selecting partition combinations on graphs without natural cut directions. We demonstrate CMFT on four GPUs with approximately four million p-bits, reaching comparable energy densities up to 15 times faster than a single-GPU implementation of the full graph. By coupling locally evolving clusters through programmable effective biases, CMFT provides a route to extreme-scale Ising machines on both analog and digital hardware.
Related
- Solving Large-Scale Parameter Inference in Controllable Nuclear Fusion via Coherent Ising Machines and Block Coordinate Descent
- Performance analysis of classical adiabatic annealing on Ising machines
- Reducing hyperparameter sensitivity in measurement-feedback based Ising machines
- Dynamic distortion from discretized coupling feedback in soft-spin Ising machines
- Evaluating the solution performance of the augmented Lagrangian function on Ising machines
BibTeX
@misc{zhang2026extremescale,
title = {{Extreme-Scale Ising Machines with Cluster Mean-Field Theory}},
author = {Xiuqi Zhang and Shuvro Chowdhury and Shaila Niazi and Christian Z. Pratt and Navid Anjum Aadit and Kerem Y. Camsari},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.22645},
eprint = {2609.22645},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.22645},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Extreme-Scale Ising Machines with Cluster Mean-Field Theory
AU - Xiuqi Zhang
AU - Shuvro Chowdhury
AU - Shaila Niazi
AU - Christian Z. Pratt
AU - Navid Anjum Aadit
AU - Kerem Y. Camsari
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.22645
UR - https://arxiv.org/abs/2609.22645
AB - Scaling analog and digital Ising machines to larger problems requires overcoming finite device capacity and the cost of communication between devices. We present cluster mean-field theory (CMFT), a framework that partitions a large interaction graph into clusters sized to fit available hardware. Each cluster performs local updates independently, while interactions across cluster boundaries enter through periodically updated effective biases computed from boundary-spin averages. To reduce the error introduced by fixed cluster boundaries, we introduce dynamic partitioning, which cycles through multiple partitions so that interactions approximated by mean fields at one stage can act through instantaneous spins at another. On three-dimensional spin glasses and planted Pegasus instances, dynamic CMFT exhibits power-law decay of residual energy over sweep budgets. This shows that solution quality continues to improve with computational effort despite the mean-field approximation. An automated graph partitioner combined with a weighted recovery ratio provides a practical heuristic for selecting partition combinations on graphs without natural cut directions. We demonstrate CMFT on four GPUs with approximately four million p-bits, reaching comparable energy densities up to 15 times faster than a single-GPU implementation of the full graph. By coupling locally evolving clusters through programmable effective biases, CMFT provides a route to extreme-scale Ising machines on both analog and digital hardware.
ER -