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

IEEE Transactions on Computational Social Systems

Beyond Classical Limits: Quantum-Driven Preselection for Multi-Asset-Class Portfolio

Zhengyang Song, Lean Yu, Bo Yang, Xiaoming Zhang, Yin Ma, Kai Wen

Published Jan 1, 2026• 0 citations
Publisher

Abstract

The exponential growth of the investable universe poses a significant computational challenge for large-scale portfolio selection, rendering traditional optimization algorithms inefficient. To address this challenge, a coherent ising machine (CIM) based hybrid quantum-classical framework is proposed to optimize the asset preselection problem for multi-asset-class portfolios. In this framework, the asset preselection is first formulated as a distance maximization problem and then transformed into a quadratic unconstrained binary optimization (QUBO) form. Finally, a hybrid quantum-classical strategy is employed to address both the NP-hard structure of the problem and the precision limitations of current CIM hardware, enabling the identification of a diversified preselected asset pool via CIM for subsequent portfolio allocation. Empirical results on U.S. stocks and global commodity futures show that the proposed framework constructs diversified multi-asset-class pools and delivers strong out-of-sample performance. It solves the baseline instances at millisecond-level latency and achieves a favorable solution-quality-runtime tradeoff relative to the benchmark algorithms. These findings demonstrate the practical potential of CIM-based optimization for large-scale portfolio preselection.

DOI
10.1109/tcss.2026.3723670

Related papers

A Quantum Approximate Optimization Algorithm (QAOA) Approach to Portfolio Optimization

Yifei Huang

Portfolio optimization, a core combinatorial optimization problem in quantitative finance, requires balancing return maximization and risk minimization within discrete asset sets. Classical algorithms often struggle with scalability and local optima traps when handling large-scale asset pools, limiting their practicality in dynamic markets. This study proposes a complete Quantum Approximate Optimization Algorithm (QAOA) workflow for portfolio optimization, from problem formalization to experimental validation. We first transform the portfolio optimization problem into a Quadratic Unconstrained Binary Optimization (QUBO) framework, then design a parameterized QAOA circuit with adjustable layers to balance expressiveness and noise resilience. We propose a methodology using Qiskit for quantum simulation and classical optimizers for parameter tuning, we propose implementing the algorithm on real-world S&P 500 stock data. We could also analyze the impact of quantum hardware noise and propose mitigation strategies, providing actionable insights for Noisy intermediate-scale quantum computing (NISQ)-era implementations. This work bridges quantum algorithm design with real-world financial applications, offering a theoretical framework for quantum-driven portfolio optimization.

Exploring Science Academic Conference Series•Jul 30, 2026•0 citations

Quantum Annealing for Dynamic Portfolio Optimization under Realistic Transaction Costs

Escolástico Sánchez-Martínez, Senaida Hernandez Santana, Ventura Sarasa Laborda +3 more

This paper investigates and compares quantum and classical investment strategies for portfolio construction under realistic trading and allocation constraints. The study considers a multi-period portfolio re-balancing setting in which asset weights are subject to lower and upper bounds, class-level exposure restrictions, and turnover limitations. On the quantum side, the portfolio selection problem is reformulated as a quadratic unconstrained binary optimization (QUBO) model through a finite binary encoding of asset weights, and subsequently embedded into a constrained quadratic model (CQM) solved by a hybrid quantum-classical optimization workflow. On the classical side, benchmark allocation strategies are introduced to provide an economically meaningful reference for performance assessment. The proposed framework jointly evaluates portfolio efficiency from two complementary perspectives: business performance, measured through return, volatility, and a measure of the trade-off between them, the Sharpe ratio; and computational performance, assessed through solver structure, constraint handling, and hardware execution characteristics. The resulting comparison provides a rigorous basis for understanding the practical role of quantum annealing in constrained portfolio optimization, while clarifying both its modeling advantages and its current implementation limitations relative to established classical approaches.

quant-pharXiv:2607.03218•Jul 6, 2026•0 citations

Portfolio Optimization using Accelerated Quantum Simulation

Gaurang Belekar, Manish Modani, Anindita Banerjee

Identifying the most suitable stocks for Portfolio stands as a critical challenge in the financial sector especially for a larger pool of assets. Quantum algorithms have emerged as a promising solution to the inherent challenges and offering the potential for exponential speedups in solving complex optimization problems. These algorithms are particularly suited for navigating the vast solution spaces associated with large asset pools, thus enabling more effective diversification strategies. Within this context, we explore the application of CUDA Quantum, an advanced quantum simulation framework and a quantum ready platform. Our study demonstrates that when applied to a set of 15 assets and more, CUDA Quantum significantly outperforms traditional CPU-based simulations in terms of computational speed and increases as the number of assets increase, highlighting the advantages of GPU acceleration in processing complex quantum algorithms. By extending our implementation to larger asset pools upto 30 assets, we provide empirical evidence of the substantial performance gains achievable with CUDA Quantum. Our findings illustrate the transformative potential of integrating quantum algorithms and GPU computing in addressing the challenge of portfolio optimization.

International Conference on Trends in Quantum Computing and Emerging Business Technologies (TQCEBT)•Mar 22, 2024•0 citations

Block Coordinate Descent for Dynamic Portfolio Optimization on Finite-Precision Coherent Ising Machines

Keming He, Yuehan Zhang, Hongshun Yao +2 more

Coherent Ising machines (CIMs) have emerged as specialized quantum hardware for large-scale combinatorial optimization. However, for large instances that remain challenging for classical methods, some platforms support only finite-precision inputs, and the required scaling and quantization can degrade solution quality. Dynamic portfolio optimization (DPO) can be formulated as a quadratic unconstrained binary optimization (QUBO) problem, but large instances are especially vulnerable to precision loss under global scaling. We propose a block coordinate descent method that decomposes the DPO model along the time dimension and iteratively solves compact time-block subproblems on the device. Experiments on finite-precision CIM hardware show that the method enables these instances to be solved under hardware precision limits, yields portfolios competitive with classical benchmark solvers, and reduces runtime through fast CIM solution of the resulting subproblems. These results demonstrate the promise of finite-precision CIMs as a practical and scalable approach to structured large-scale combinatorial optimization.

quant-pharXiv:2603.23200•Mar 25, 2026•0 citations

Solving Multiple Discretization Portfolio Optimization Problem with Quantum-Classical Hybrid Algorithms

Haijing Wei, Yanbo J. Wang, Haoxiang Yang +12 more

Portfolio optimization (PO) plays a central role in finance, encompassing both resource allocation and risk management. While numerous algorithms have been developed for continuous-variable PO, the inclusion of large, indivisible assets such as real estate transforms the task into a combinatorial optimization problem, which is NP-hard for classical computers. This study explores using quantum-classical hybrid algorithms for multi-discretized portfolio optimization, addressing the needs of real-world applications. We propose a flexible theoretical model that can solve PO problems with various types of discrete assets and constraints, which is compatible with both quantum and classical computing paradigms. Then, we use the D-Wave quantum processor and classical solver to determine optimal investment strategies. Experimental results indicate that the hybrid quantum algorithm outperforms both classical computing methods and quantum annealing algorithms in terms of return and risk management. This research enhances quantum computing applications in finance and provides insights into future algorithm design.

Computational Economics•Jul 31, 2025•2 citations

Quantum-Enhanced Stock Portfolio Optimization Using Normalizing Flows and QAOA: Design, Implementation, and Validation

Chirag Solanki, Parthiv Somaiya, Payal Varangaonkar

This paper presents a hybrid quantum-classical portfolio optimization pipeline that integrates a Quantum Normalizing Flow (QNF) scenario generator with a Quantum Approximate Optimization Algorithm (QAOA) binary asset selector. The objective is to improve discrete stock selection under uncertainty by capturing nonlinear return dependencies with QNF and solving a Quadratic Unconstrained Binary Optimization (QUBO) formulation via QAOA. The methodology is validated on ten large-cap U.S. assets and compared against three classical baselines (Equal Weight, Minimum Variance, Maximum Sharpe), using risk-adjusted metrics (annualized return, volatility, Sharpe ratio) and Monte Carlo statistical testing. Key contributions: 1) A modular QNF-QAOA hybrid optimizer for robust asset selection. 2) A rigorously reproducible evaluation pipeline with quantitative performance and statistical significance analysis. 3) Empirical evidence of competitive riskadjusted returns and conservative drawdown behavior.

International Conference on Intelligent and Sustainable Electronics & Computing Technologies (INSECT)•May 1, 2026•0 citations
The exponential growth of the investable universe poses a significant computational challenge for large-scale portfolio selection, rendering traditional optimization algorithms inefficient. To address this challenge, a coherent ising machine (CIM) based hybrid quantum-classical framework is proposed to optimize the asset preselection problem for multi-asset-class portfolios. In this framework, the asset preselection is first formulated as a distance maximization problem and then transformed into a quadratic unconstrained binary optimization (QUBO) form. Finally, a hybrid quantum-classical strategy is employed to address both the NP-hard structure of the problem and the precision limitations of current CIM hardware, enabling the identification of a diversified preselected asset pool via CIM for subsequent portfolio allocation. Empirical results on U.S. stocks and global commodity futures show that the proposed framework constructs diversified multi-asset-class pools and delivers strong out-of-sample performance. It solves the baseline instances at millisecond-level latency and achieves a favorable solution-quality-runtime tradeoff relative to the benchmark algorithms. These findings demonstrate the practical potential of CIM-based optimization for large-scale portfolio preselection.
Portfolio optimization, a core combinatorial optimization problem in quantitative finance, requires balancing return maximization and risk minimization within discrete asset sets. Classical algorithms often struggle with scalability and local optima traps when handling large-scale asset pools, limiting their practicality in dynamic markets. This study proposes a complete Quantum Approximate Optimization Algorithm (QAOA) workflow for portfolio optimization, from problem formalization to experimental validation. We first transform the portfolio optimization problem into a Quadratic Unconstrained Binary Optimization (QUBO) framework, then design a parameterized QAOA circuit with adjustable layers to balance expressiveness and noise resilience. We propose a methodology using Qiskit for quantum simulation and classical optimizers for parameter tuning, we propose implementing the algorithm on real-world S&P 500 stock data. We could also analyze the impact of quantum hardware noise and propose mitigation strategies, providing actionable insights for Noisy intermediate-scale quantum computing (NISQ)-era implementations. This work bridges quantum algorithm design with real-world financial applications, offering a theoretical framework for quantum-driven portfolio optimization.
This paper investigates and compares quantum and classical investment strategies for portfolio construction under realistic trading and allocation constraints. The study considers a multi-period portfolio re-balancing setting in which asset weights are subject to lower and upper bounds, class-level exposure restrictions, and turnover limitations. On the quantum side, the portfolio selection problem is reformulated as a quadratic unconstrained binary optimization (QUBO) model through a finite binary encoding of asset weights, and subsequently embedded into a constrained quadratic model (CQM) solved by a hybrid quantum-classical optimization workflow. On the classical side, benchmark allocation strategies are introduced to provide an economically meaningful reference for performance assessment. The proposed framework jointly evaluates portfolio efficiency from two complementary perspectives: business performance, measured through return, volatility, and a measure of the trade-off between them, the Sharpe ratio; and computational performance, assessed through solver structure, constraint handling, and hardware execution characteristics. The resulting comparison provides a rigorous basis for understanding the practical role of quantum annealing in constrained portfolio optimization, while clarifying both its modeling advantages and its current implementation limitations relative to established classical approaches.
Identifying the most suitable stocks for Portfolio stands as a critical challenge in the financial sector especially for a larger pool of assets. Quantum algorithms have emerged as a promising solution to the inherent challenges and offering the potential for exponential speedups in solving complex optimization problems. These algorithms are particularly suited for navigating the vast solution spaces associated with large asset pools, thus enabling more effective diversification strategies. Within this context, we explore the application of CUDA Quantum, an advanced quantum simulation framework and a quantum ready platform. Our study demonstrates that when applied to a set of 15 assets and more, CUDA Quantum significantly outperforms traditional CPU-based simulations in terms of computational speed and increases as the number of assets increase, highlighting the advantages of GPU acceleration in processing complex quantum algorithms. By extending our implementation to larger asset pools upto 30 assets, we provide empirical evidence of the substantial performance gains achievable with CUDA Quantum. Our findings illustrate the transformative potential of integrating quantum algorithms and GPU computing in addressing the challenge of portfolio optimization.
Coherent Ising machines (CIMs) have emerged as specialized quantum hardware for large-scale combinatorial optimization. However, for large instances that remain challenging for classical methods, some platforms support only finite-precision inputs, and the required scaling and quantization can degrade solution quality. Dynamic portfolio optimization (DPO) can be formulated as a quadratic unconstrained binary optimization (QUBO) problem, but large instances are especially vulnerable to precision loss under global scaling. We propose a block coordinate descent method that decomposes the DPO model along the time dimension and iteratively solves compact time-block subproblems on the device. Experiments on finite-precision CIM hardware show that the method enables these instances to be solved under hardware precision limits, yields portfolios competitive with classical benchmark solvers, and reduces runtime through fast CIM solution of the resulting subproblems. These results demonstrate the promise of finite-precision CIMs as a practical and scalable approach to structured large-scale combinatorial optimization.
Portfolio optimization (PO) plays a central role in finance, encompassing both resource allocation and risk management. While numerous algorithms have been developed for continuous-variable PO, the inclusion of large, indivisible assets such as real estate transforms the task into a combinatorial optimization problem, which is NP-hard for classical computers. This study explores using quantum-classical hybrid algorithms for multi-discretized portfolio optimization, addressing the needs of real-world applications. We propose a flexible theoretical model that can solve PO problems with various types of discrete assets and constraints, which is compatible with both quantum and classical computing paradigms. Then, we use the D-Wave quantum processor and classical solver to determine optimal investment strategies. Experimental results indicate that the hybrid quantum algorithm outperforms both classical computing methods and quantum annealing algorithms in terms of return and risk management. This research enhances quantum computing applications in finance and provides insights into future algorithm design.
This paper presents a hybrid quantum-classical portfolio optimization pipeline that integrates a Quantum Normalizing Flow (QNF) scenario generator with a Quantum Approximate Optimization Algorithm (QAOA) binary asset selector. The objective is to improve discrete stock selection under uncertainty by capturing nonlinear return dependencies with QNF and solving a Quadratic Unconstrained Binary Optimization (QUBO) formulation via QAOA. The methodology is validated on ten large-cap U.S. assets and compared against three classical baselines (Equal Weight, Minimum Variance, Maximum Sharpe), using risk-adjusted metrics (annualized return, volatility, Sharpe ratio) and Monte Carlo statistical testing. Key contributions: 1) A modular QNF-QAOA hybrid optimizer for robust asset selection. 2) A rigorously reproducible evaluation pipeline with quantitative performance and statistical significance analysis. 3) Empirical evidence of competitive riskadjusted returns and conservative drawdown behavior.