Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Neuromorphic Computing and Engineering

A neuromorphic digital Ising solver with Tabu-inspired inhibitory dynamics for scalable graph optimization

Juan Núñez, Rafaella Fiorelli

Published Jul 24, 2026• 0 citations
Publisher

Abstract

Abstract We present a fully digital Ising solver for maximum-cut implemented on a low-cost Artix-7 field-programmable gate array (FPGA), where tabu-inspired inhibitory dynamics are realized as a compact, Block RAM-resident short-term memory. The solver operates in deterministic fixed-point arithmetic and scales up to the on-chip limit of N = 850 spins across multiple graph sizes and sparsity regimes. We benchmark the proposed Tabu-based dynamics against (i) a parallel Hopfield-network update rule implemented on the same FPGA, and (ii) a CPU-based quantum approximate optimization algorithm (QAOA) reference used as a fixed variational baseline for solution quality under a prescribed budget. Across the tested graph families, the Tabu-enhanced solver reaches higher cut values with reduced run-to-run dispersion than purely deterministic descent. Quantitatively, field-aligned warm start improves the median normalized cut of TS by 0.014 and reduces solve time by 0.18 ms at N = 100 , ρ = 20 % . At the largest tested size ( N = 850 ), TS improves the median normalized cut over the fixed-budget CPU-QAOA reference by 0.014–0.016, with median CPU–FPGA solve-time differences of 640–720 ms under the reported protocol. The N = 850 FPGA implementation closes timing at 100 MHz, demonstrating that short-term inhibitory memory can be embedded in a fully digital Ising network without relying on stochasticity, analog variability, or annealing schedules.

DOI
10.1088/2634-4386/ae9006

Related papers

Restoring sparsity in potts machines via mean-field constraints

Kevin Callahan-Coray, K. Lee, Kyle Jiang +1 more

Abstract Ising machines and related probabilistic hardware have emerged as promising platforms for NP-hard optimization and sampling. However, many practical problems involve constraints that induce dense or all-to-all couplings, undermining scalability and hardware efficiency. We address this constraint-induced density through two complementary approaches. First, we introduce a hardware-aware native formulation for multi-state probabilistic digits (p-dits) that avoids the locally dense intra-variable couplings required by binary Ising encodings. We validate p-dit dynamics by reproducing known critical behavior of the 2D Potts model. Second, we propose mean-field constraints (MFC), a hybrid scheme that replaces dense pairwise constraint couplings with dynamically updated single-node biases. Applied to balanced graph partitioning, MFC achieves solution quality comparable to exact all-to-all constraint formulations while dramatically reducing graph density. Finally, we demonstrate the practical impact of restored sparsity through an FPGA implementation. In comparisons using FPGA kernel time and CPU solver-loop time, and excluding the current prototype’s host-device schedule-transfer overhead, the FPGA reaches the 50% success threshold more than an order of magnitude faster than the CPU probabilistic solvers and more than two orders of magnitude faster than the Tabu Ising baseline. Together, these results outline a pathway for scaling constrained optimization on probabilistic hardware.

arXiv:2602.04200•npj Unconventional Computing•Sep 2, 2026•3 citations

Design and FPGA Realization of QUBO hardware accelerator for MAX-CUT problem

Maciej Lewandowski, Piotr Dudek

Combinatorial optimization problems are fundamental to many industrial and scientific applications, but are often NP-hard, requiring heuristic and probabilistic approaches to find good solutions fast. This paper presents the first hardware implementation of the neuromorphic-inspired NeuroSA algorithm, for accelerating Quadratic Unconstrained Binary Optimization (QUBO) problems. We show that it is possible to run a standard QUBO benchmark problem with 800 variables on a low-cost FPGA with a clock speed of approximately 144 MHz, which converges to the same solution value found by the numerical solver.

IEEE International Symposium on Circuits and Systems proceedings•May 25, 2025•2 citations

Ising Model Optimization Problems on a FPGA Accelerated Restricted Boltzmann Machine

Saavan Patel, Lili Chen, Philip Canoza +1 more

Abstract In this work we demonstrate usage of the Restricted Boltzmann Machine (RBM) as a stochastic neural network capable of solving NP-Hard Combinatorial Optimization problems efficiently. By mapping the RBM onto a reconfigurable Field Programmable Gate Array (FPGA), we can effectively hardware accelerate the RBM's stochastic sampling algorithm. We benchmark the RBM against the DWave 2000Q Quantum Adiabatic Computer and the Optical Coherent Ising Machine on two such optimization problems: the MAX-CUT problem and the Sherrington-Kirkpatrick (SK) spin glass. The hardware accelerated RBM shows asymptotic scaling either similar or better than these other accelerators. This leads to 10 7 x and 10 5 x time to solution improvement compared to the DWave 2000Q on the MAX-CUT and SK problems respectively, along with a 150x and 1000x improvement compared to the Coherent Ising Machine annealer on those problems. By utilizing commodity hardware running at room temperature, the RBM shows potential for immediate and scalable use.

arXiv:2008.04436•Dec 15, 2020•40 citations

High-Parallel FPGA-Based Discrete Simulated Bifurcation for Large-Scale Optimization

Fabrizio Orlando, Deborah Volpe, Orlandi, Giacomo +3 more

Combinatorial Optimization (CO) problems exhibit exponential complexity, making their resolution challenging. Simulated Adiabatic Bifurcation (aSB) is a quantum-inspired algorithm to obtain approximate solutions to largescale CO problems written in the Ising form. It explores the solution space by emulating the adiabatic evolution of a network of Kerr-nonlinear parametric oscillators (KPOs), where each oscillator represents a variable in the problem. The optimal solution corresponds to the ground state of this system. A key advantage of this approach is the possibility of updating multiple variables simultaneously, making it particularly suited for hardware implementation. To enhance solution quality and convergence speed, variations of the algorithm have been proposed in the literature, including ballistic (bSB), discrete (dSB), and thermal (HbSB) versions. In this work, we have comprehensively analyzed dSB, bSB, and HbSB using dedicated software models, evaluating the feasibility of using a fixed-point representation for hardware implementation. We then present an opensource hardware architecture implementing the dSB algorithm for Field-Programmable Gate Arrays (FPGAs). The design allows users to adjust the degree of algorithmic parallelization based on their specific requirements. A proof-of-concept implementation that solves 256-variable problems was achieved on an AMD Kria KV260 SoM, a low-tier FPGA, validated using well-known max-cut and knapsack problems.

arXiv:2510.12407•Oct 14, 2025•2 citations

A 1.87-TELOPS/W 3-D Ising Machine for Accelerated Quantum Monte Carlo With Reconfigurability Using CMOS p-Bits

Ming-Che Li, Archisman Ghosh, Risi Jaiswal +2 more

Qubit-based quantum annealing processors operate at ultra-low temperatures (15 mK) requiring enormous cooling energy. Consequently, there is a lot of interest in quantum Monte Carlo (QMC) algorithms that can be used to emulate quantum computing on classical machines. However, prior classical emulators implemented on CPUs, GPUs, or field-programmable gate arrays (FPGAs) are orders of magnitude slower than quantum annealers highlighting the need for dedicated application-specific integrated circuits (ASICs). Over the past decade, many ASICs have been developed, including general Ising solvers, typically implemented as digital system on chip (SoCs) targeting highly complex graphs, and specific Ising solvers, which address relatively simple Ising graphs using analog computing paradigms. General solvers often suffer from high power consumption, while specific solvers lack reconfigurability. This work presents, for the first time, a QMC algorithm implemented on a 3-D Ising computer fabricated in commercial 65-nm CMOS technology. Specifically, it solves a quantum problem in which a transverse field Ising (TFI) Hamiltonian is applied to an unconventional octagonal lattice. The reconfigurable integrated circuit (IC) also supports traditional 2-D Ising models for solving combinatorial optimization problems (COPs), offering better reconfigurability than prior specific Ising solvers. This IC achieved an energy-to-solution of$41.9~\mu $J at 1 MHz and$16.65~\mu $J at 10 MHz at 0.5 V for the QMC problem. Compared to previous FPGA implementations, this is a three-order-of-magnitude improvement and a six-order-of-magnitude improvement over CPU-based solutions. This article also proposes a new energy efficiency metric,$E_{\text {ELOP}}$(Energy per ELementary OPeration), to better benchmark p-bit-based Ising solvers. Our IC achieves an$E_{\text {ELOP}}$of 1350 fJ at 1 MHz and 535 fJ at 10 MHz (measured at 0.5 V) for solving the QMC problem, which are also equivalent to the energy efficiency of 0.74 and 1.87 TELOPS/W (ELementary OPeration per Second per Watt).

IEEE Journal of Solid-State Circuits•Jun 17, 2026•0 citations

Engineering Discrete Simulated Bifurcation for an FPGA Digital Ising Machine

Fabrizio Orlando, Deborah Volpe, Giacomo Orlandi +3 more

Simulated Adiabatic Bifurcation (aSB) is a quantum-inspired algorithm that provides approximate solutions for large-scale optimization problems using the Ising model. It emulates the quantum adiabatic evolution of a network of non-linear Kerr oscillators on classical platforms. These oscillators undergo bifurcation, where each branch corresponds to a spin state, with the network creating an energy imbalance to deter-mine the optimal solution. This approach is highly parallelizable, making it suitable for implementation on GPUs and FPGAs. However, classical emulation introduces analog errors, potentially compromising performance. To address this, alternative methods like ballistic (bSB) and discrete (dSB) evolutions were developed, with the ballistic approach further enhanced by thermal fluctuations (HbSB). Our comprehensive analysis of bSB, dSB, and HbSB using benchmark Max-Cut and knapsack problems aimed to identify the best balance of speed, area, and accuracy. We found that fixed-point number representation allows more algorithm iterations within the same timeframe as floating-point representation. The proposed FPGA architecture implements d$S$B and its heated version, leveraging high parallelizability and efficient memory or-ganization. This design supports larger coefficients and improves problem-solving efficiency without multipliers for matrix-vector evaluation. Challenges include mitigating data dependency between matrix-vector multiplication and time evolution, and developing a preconditioning method for adapting problem coefficients, making the architecture suitable for real-world applications and potential ASIC release.

IEEE International Conference on Quantum Computing and Engineering (QCE)•Sep 15, 2024•1 citation
Abstract We present a fully digital Ising solver for maximum-cut implemented on a low-cost Artix-7 field-programmable gate array (FPGA), where tabu-inspired inhibitory dynamics are realized as a compact, Block RAM-resident short-term memory. The solver operates in deterministic fixed-point arithmetic and scales up to the on-chip limit of N = 850 spins across multiple graph sizes and sparsity regimes. We benchmark the proposed Tabu-based dynamics against (i) a parallel Hopfield-network update rule implemented on the same FPGA, and (ii) a CPU-based quantum approximate optimization algorithm (QAOA) reference used as a fixed variational baseline for solution quality under a prescribed budget. Across the tested graph families, the Tabu-enhanced solver reaches higher cut values with reduced run-to-run dispersion than purely deterministic descent. Quantitatively, field-aligned warm start improves the median normalized cut of TS by 0.014 and reduces solve time by 0.18 ms at N = 100 , ρ = 20 % . At the largest tested size ( N = 850 ), TS improves the median normalized cut over the fixed-budget CPU-QAOA reference by 0.014–0.016, with median CPU–FPGA solve-time differences of 640–720 ms under the reported protocol. The N = 850 FPGA implementation closes timing at 100 MHz, demonstrating that short-term inhibitory memory can be embedded in a fully digital Ising network without relying on stochasticity, analog variability, or annealing schedules.
Abstract Ising machines and related probabilistic hardware have emerged as promising platforms for NP-hard optimization and sampling. However, many practical problems involve constraints that induce dense or all-to-all couplings, undermining scalability and hardware efficiency. We address this constraint-induced density through two complementary approaches. First, we introduce a hardware-aware native formulation for multi-state probabilistic digits (p-dits) that avoids the locally dense intra-variable couplings required by binary Ising encodings. We validate p-dit dynamics by reproducing known critical behavior of the 2D Potts model. Second, we propose mean-field constraints (MFC), a hybrid scheme that replaces dense pairwise constraint couplings with dynamically updated single-node biases. Applied to balanced graph partitioning, MFC achieves solution quality comparable to exact all-to-all constraint formulations while dramatically reducing graph density. Finally, we demonstrate the practical impact of restored sparsity through an FPGA implementation. In comparisons using FPGA kernel time and CPU solver-loop time, and excluding the current prototype’s host-device schedule-transfer overhead, the FPGA reaches the 50% success threshold more than an order of magnitude faster than the CPU probabilistic solvers and more than two orders of magnitude faster than the Tabu Ising baseline. Together, these results outline a pathway for scaling constrained optimization on probabilistic hardware.
Combinatorial optimization problems are fundamental to many industrial and scientific applications, but are often NP-hard, requiring heuristic and probabilistic approaches to find good solutions fast. This paper presents the first hardware implementation of the neuromorphic-inspired NeuroSA algorithm, for accelerating Quadratic Unconstrained Binary Optimization (QUBO) problems. We show that it is possible to run a standard QUBO benchmark problem with 800 variables on a low-cost FPGA with a clock speed of approximately 144 MHz, which converges to the same solution value found by the numerical solver.
Abstract In this work we demonstrate usage of the Restricted Boltzmann Machine (RBM) as a stochastic neural network capable of solving NP-Hard Combinatorial Optimization problems efficiently. By mapping the RBM onto a reconfigurable Field Programmable Gate Array (FPGA), we can effectively hardware accelerate the RBM's stochastic sampling algorithm. We benchmark the RBM against the DWave 2000Q Quantum Adiabatic Computer and the Optical Coherent Ising Machine on two such optimization problems: the MAX-CUT problem and the Sherrington-Kirkpatrick (SK) spin glass. The hardware accelerated RBM shows asymptotic scaling either similar or better than these other accelerators. This leads to 10 7 x and 10 5 x time to solution improvement compared to the DWave 2000Q on the MAX-CUT and SK problems respectively, along with a 150x and 1000x improvement compared to the Coherent Ising Machine annealer on those problems. By utilizing commodity hardware running at room temperature, the RBM shows potential for immediate and scalable use.
Combinatorial Optimization (CO) problems exhibit exponential complexity, making their resolution challenging. Simulated Adiabatic Bifurcation (aSB) is a quantum-inspired algorithm to obtain approximate solutions to largescale CO problems written in the Ising form. It explores the solution space by emulating the adiabatic evolution of a network of Kerr-nonlinear parametric oscillators (KPOs), where each oscillator represents a variable in the problem. The optimal solution corresponds to the ground state of this system. A key advantage of this approach is the possibility of updating multiple variables simultaneously, making it particularly suited for hardware implementation. To enhance solution quality and convergence speed, variations of the algorithm have been proposed in the literature, including ballistic (bSB), discrete (dSB), and thermal (HbSB) versions. In this work, we have comprehensively analyzed dSB, bSB, and HbSB using dedicated software models, evaluating the feasibility of using a fixed-point representation for hardware implementation. We then present an opensource hardware architecture implementing the dSB algorithm for Field-Programmable Gate Arrays (FPGAs). The design allows users to adjust the degree of algorithmic parallelization based on their specific requirements. A proof-of-concept implementation that solves 256-variable problems was achieved on an AMD Kria KV260 SoM, a low-tier FPGA, validated using well-known max-cut and knapsack problems.
Qubit-based quantum annealing processors operate at ultra-low temperatures (15 mK) requiring enormous cooling energy. Consequently, there is a lot of interest in quantum Monte Carlo (QMC) algorithms that can be used to emulate quantum computing on classical machines. However, prior classical emulators implemented on CPUs, GPUs, or field-programmable gate arrays (FPGAs) are orders of magnitude slower than quantum annealers highlighting the need for dedicated application-specific integrated circuits (ASICs). Over the past decade, many ASICs have been developed, including general Ising solvers, typically implemented as digital system on chip (SoCs) targeting highly complex graphs, and specific Ising solvers, which address relatively simple Ising graphs using analog computing paradigms. General solvers often suffer from high power consumption, while specific solvers lack reconfigurability. This work presents, for the first time, a QMC algorithm implemented on a 3-D Ising computer fabricated in commercial 65-nm CMOS technology. Specifically, it solves a quantum problem in which a transverse field Ising (TFI) Hamiltonian is applied to an unconventional octagonal lattice. The reconfigurable integrated circuit (IC) also supports traditional 2-D Ising models for solving combinatorial optimization problems (COPs), offering better reconfigurability than prior specific Ising solvers. This IC achieved an energy-to-solution of$41.9~\mu $J at 1 MHz and$16.65~\mu Jat10MHzat0.5VfortheQMCproblem.ComparedtopreviousFPGAimplementations,thisisathree−order−of−magnitudeimprovementandasix−order−of−magnitudeimprovementoverCPU−basedsolutions.Thisarticlealsoproposesanewenergyefficiencymetric,J at 10 MHz at 0.5 V for the QMC problem. Compared to previous FPGA implementations, this is a three-order-of-magnitude improvement and a six-order-of-magnitude improvement over CPU-based solutions. This article also proposes a new energy efficiency metric,Jat10MHzat0.5VfortheQMCproblem.ComparedtopreviousFPGAimplementations,thisisathree−order−of−magnitudeimprovementandasix−order−of−magnitudeimprovementoverCPU−basedsolutions.Thisarticlealsoproposesanewenergyefficiencymetric,E_{\text {ELOP}}(EnergyperELementaryOPeration),tobetterbenchmarkp−bit−basedIsingsolvers.OurICachievesan(Energy per ELementary OPeration), to better benchmark p-bit-based Ising solvers. Our IC achieves an(EnergyperELementaryOPeration),tobetterbenchmarkp−bit−basedIsingsolvers.OurICachievesanE_{\text {ELOP}}$of 1350 fJ at 1 MHz and 535 fJ at 10 MHz (measured at 0.5 V) for solving the QMC problem, which are also equivalent to the energy efficiency of 0.74 and 1.87 TELOPS/W (ELementary OPeration per Second per Watt).
Simulated Adiabatic Bifurcation (aSB) is a quantum-inspired algorithm that provides approximate solutions for large-scale optimization problems using the Ising model. It emulates the quantum adiabatic evolution of a network of non-linear Kerr oscillators on classical platforms. These oscillators undergo bifurcation, where each branch corresponds to a spin state, with the network creating an energy imbalance to deter-mine the optimal solution. This approach is highly parallelizable, making it suitable for implementation on GPUs and FPGAs. However, classical emulation introduces analog errors, potentially compromising performance. To address this, alternative methods like ballistic (bSB) and discrete (dSB) evolutions were developed, with the ballistic approach further enhanced by thermal fluctuations (HbSB). Our comprehensive analysis of bSB, dSB, and HbSB using benchmark Max-Cut and knapsack problems aimed to identify the best balance of speed, area, and accuracy. We found that fixed-point number representation allows more algorithm iterations within the same timeframe as floating-point representation. The proposed FPGA architecture implements dSSSB and its heated version, leveraging high parallelizability and efficient memory or-ganization. This design supports larger coefficients and improves problem-solving efficiency without multipliers for matrix-vector evaluation. Challenges include mitigating data dependency between matrix-vector multiplication and time evolution, and developing a preconditioning method for adapting problem coefficients, making the architecture suitable for real-world applications and potential ASIC release.