CYCU Quantum Finance Research Published in Nature Portfolio Journal, Partnering with Leading Global Teams to Forge FinTech Breakthroughs

New Strategy Overcomes Traditional Financial Model Limitations to Significantly Boost Returns and Slash Transaction Costs

【Taoyuan News】

Quantum computing technology has achieved a landmark breakthrough in the field of financial technology (FinTech). An international research team led by Assistant Professor Yen-Jui Chang of the Quantum Information Center at Chung Yuan Christian University (CYCU) has successfully developed a cutting-edge portfolio optimization technique based on “Quantum Stochastic Walks (QSW).” The study introduces an innovative “Smart 1/N” asset allocation strategy that substantially enhances investment returns while mitigating transaction costs. In recognition of its pioneering academic merit, the research has been published in npj Unconventional Computing, a premier international journal under the prestigious Nature Portfolio.

This breakthrough seamlessly fuses the academic prowess of Taiwan and the United Kingdom, bolstered by technical support from artificial intelligence and semiconductor giant NVIDIA. The core research team includes Professor Yen-Jui Chang, Professor Ching-Ray Chang, and their research group from CYCU, alongside Dr. Kuan-Cheng Chen from Imperial College London, demonstrating CYCU’s profound strength in driving global academic collaborations.

Pioneering the “Smart 1/N” Strategy: Making Investments Smarter and More Cost-Efficient

While traditional financial portfolio models—such as Markowitz’s Mean-Variance Model—are theoretically robust, they frequently grapple with “estimation errors” and “over-trading” in real-world applications, which results in heavy brokerage fees that erode net profits. On the flip side, the classic and straightforward “equal-weighting” or “1/N” strategy is inherently stable but completely fails to leverage real-time market data to optimize returns.

The QSW algorithm developed by the CYCU team ingeniously synthesizes the interference effects of quantum mechanics with the structural stability of classical random walks, creating a proprietary “Smart 1/N” strategy.

“Our model maintains the high risk-diversification benefits typical of a 1/N portfolio, while simultaneously utilizing quantum algorithms to detect and micro-adjust for latent profitable opportunities in the market,” explained Professor Yen-Jui Chang. “It is essentially adding an intelligent navigation system on top of an ultra-secure foundation.”

Stellar Empirical Data: Surging Profitability Metrics, Plummeting Transaction Costs

The research team backtested the “Smart 1/N” strategy against S&P 500 index constituents over an extensive 34-year period (1990–2024). The empirical results revealed extraordinary performance across all key metrics:

• Enhanced Profitability: In terms of the Sharpe Ratio—the primary metric for risk-adjusted returns—the QSW model achieved a score of 0.979, vastly outperforming the traditional model’s 0.480 and the standard 1/N strategy’s 0.742.

• Superior Drawdown Resilience: The model’s Calmar Ratio (which measures return relative to maximum drawdown) surged to 0.63, significantly higher than the traditional model’s 0.28, proving its exceptional resilience during severe market crashes.

• Drastically Lowered Transaction Costs: Addressing a primary pain point for institutional investors, the new model effectively suppresses redundant, high-frequency trading. By keeping portfolio turnover at an exceptionally low level, it saves massive amounts in transaction and brokerage fees compared to traditional models.

A Model for Global Industry-Academia Synergies: Powered by NVIDIA

A standout highlight of this initiative is the tight-knit global synergy between industry and academia. The highly intricate, quantum-inspired heuristics used in the study were fully backed by the NVIDIA AI Technology Center (NVAITC). Utilizing NVIDIA’s advanced GPU acceleration technologies alongside the CUDA-Q platform architecture, the team executed massive financial network simulations within a remarkably compressed timeframe, validating the immense potential of quantum algorithms in processing big-data financial complexities.

Furthermore, the collaboration with Imperial College London deeply integrated quantum engineering with financial network theory, injecting a global perspective into Taiwan’s homegrown quantum finance research ecosystem.

CYCU stated that this publication in a Nature portfolio journal serves as concrete validation that the university’s research capabilities in “Intelligent Computing” and “Quantum Information” have reached elite international standards. It offers tangible, practical solutions for the next generation of FinTech architecture. Moving forward, CYCU will continue to spearhead cross-disciplinary and transnational collaborations to nurture top-tier talent equipped with future-ready technological capabilities.

【Publication Information】

• Paper Title: Quantum stochastic walks for portfolio optimization: theory and implementation on financial networks

• Journal: npj Unconventional Computing (Nature Portfolio)

• Full Research Article Link: https://www.nature.com/articles/s44335-025-00050-4