Nejla Essaddi, Imen Ziadi, and Mongi Besbes, affiliated with the University of Tunis El Manar, SUP’COM, and the Higher Institute of Information and Communication Technologies at the University of Carthage, set out to answer a question that has hovered over the quantum machine learning community for a decade: does quantum computing offer any genuine advantage for practical classification problems, or is it all theoretical promise? Rather than testing on toy datasets, the team turned to the Mediterranean, one of the most tectonically complicated regions on Earth, where the African and Eurasian plates collide in a patchwork of subduction zones, strike-slip faults, and continental collision. They drew their data from the United States Geological Survey catalog, a publicly accessible record of global seismicity, and framed the task as a binary classification problem: does an event exceed local magnitude 4.0, the threshold above which earthquakes begin to pose genuine hazards?

The scale of the benchmark is what distinguishes this study from much of the quantum machine learning literature. The researchers evaluated 261 distinct experimental configurations, systematically varying the temporal windows of seismic features fed into each model, the dimensionality of those feature vectors, and the choice of classifier. On the classical side, they tested logistic regression, the workhorse of interpretable statistics; random forests, ensembles of decision trees renowned for robustness; and support vector machines, which separate data classes by finding optimal boundaries in transformed feature spaces. On the quantum side, they deployed two fundamentally different architectures: the variational quantum classifier, a hybrid quantum-classical circuit whose parameters are tuned by a classical optimizer, and the quantum kernel support vector machine, which computes distances between data points in an exponentially large quantum feature space and feeds those similarities to a classical SVM.

The technical distinction between these two quantum approaches matters enormously for interpreting the results. Variational quantum classifiers function like quantum neural networks: data is encoded into qubit states through parameterized rotation gates, a measurement produces a prediction, and the parameters are iteratively adjusted to minimize a loss function. Quantum kernel methods, by contrast, bypass training of the quantum circuit altogether. Instead, each pair of data points is loaded into a quantum circuit whose evolution depends on the data values, and the overlap between the resulting quantum states—a quantity requiring exponentially many classical operations to compute exactly—serves as a kernel in a classical support vector machine. When the encoding circuit is chosen well, this kernel can capture patterns that are classically hard to represent, which is precisely the kind of advantage quantum machine learning theorists have been hunting for.

The headline finding is nuanced but striking. Across the full Mediterranean dataset, the classical random forest remained the most dependable global performer, reaching accuracies as high as 0.833. But in specific tectonic settings—particular zones combined with particular temporal histories of seismic activity—the entangled quantum kernel SVM achieved dramatically better results. Its peak accuracy reached 0.947, a 28.5 percent relative improvement over the best classical result in that same zone-history combination, achieved with only six qubits. In an era when quantum hardware is noisy, limited, and expensive, the fact that a six-qubit model could dominate a tuned random forest on any slice of real geophysical data is a genuinely remarkable result.

The authors, however, are careful not to oversell the finding, and their statistical honesty is part of what makes the study valuable. That peak performance did not reach statistical significance in their pre-specified fixed-model comparison; after applying the Holm–Bonferroni correction for multiple comparisons, the adjusted p-value was 1.0. The reasons are practical rather than mysterious: test sets in narrow tectonic zones are small, and the standout quantum result emerged from post-hoc model selection—examining many configurations and highlighting the best one—which inherently inflates apparent performance. The team explicitly frames these results as exploratory, a signal worth pursuing rather than a settled proof of quantum advantage. This kind of methodological transparency is rare in a field often criticized for hype, and it sets a benchmark for how quantum machine learning claims should be reported.

The study also delivered a cautionary tale about variational quantum classifiers. While quantum kernel methods thrived in favorable geometries, the VQC models suffered severe training instability as qubit counts increased—a phenomenon well known in the quantum computing literature as the barren plateau problem, where gradients of the loss function vanish exponentially with circuit size, leaving optimizers wandering a nearly flat landscape with no useful direction. McClean and colleagues first characterized this pathology in 2018, and the Mediterranean earthquake data confirms it in practice: pushing the variational approach to higher qubit counts degraded rather than improved results. Intriguingly, however, the VQC showed unexpected niche competitiveness in data-scarce regimes, hinting that different quantum architectures may suit different data availability conditions—a finding with real implications for seismically active but poorly instrumented regions.

What does it mean for a quantum feature space to align with geophysical data geometry? The Mediterranean’s seismotectonic zones produce feature distributions shaped by interacting fault systems, depth-dependent attenuation, and regional magnitude scales. When the embedding circuit entangles features in a way that mirrors these physical correlations, the quantum kernel can draw decision boundaries that classical kernels approximate only crudely. The 0.947 accuracy in specific configurations suggests that, at least locally, the structure of earthquake catalog features resonates with the expressivity of a modest quantum circuit. The authors argue this points toward targeted, hybrid quantum-classical early warning systems, in which classical models handle most of the workload and quantum kernels are deployed selectively where their strengths apply.

The practical roadmap emerging from this research is one of selective hybridization rather than wholesale replacement. Earthquake early warning is a domain where seconds matter and false alarms carry real economic and social costs, so any improvement in classification reliability, even in narrow regimes, is consequential. The Tunisian team’s work demonstrates that quantum kernels deserve a place in the toolkit—not as a universal solution, but as a specialized instrument whose deployment should be guided by the geometry of the data and the tectonic character of the region. As quantum hardware matures and qubit counts grow beyond the six used here, the boundary between classical and quantum competitiveness will inevitably shift. For now, the Mediterranean has provided the testing ground where quantum machine learning took a measurable, statistically honest step from theory toward the seismic frontier.

Subject of Research: Benchmarking quantum kernel machine learning against classical classifiers for earthquake magnitude classification in the Mediterranean region.

Article Title: Harnessing quantum kernels for robust earthquake classification: a Mediterranean case study

Article References: Essaddi, N., Ziadi, I., & Besbes, M. (2026). Harnessing quantum kernels for robust earthquake classification: a Mediterranean case study. Quantum Machine Intelligence, 8(2), Article 102. https://doi.org/10.1007/s42484-026-00443-z

Image Credits: AI Generated

DOI: 10.1007/s42484-026-00443-z

Keywords: quantum machine learning, earthquake classification, quantum kernel SVM, variational quantum classifier, Mediterranean seismicity, USGS catalog, random forest, support vector machine, early warning systems, seismology, six-qubit circuits, Holm-Bonferroni correction

Optimization is a cornerstone of numerous scientific and industrial endeavors, from financial portfolio management to machine learning and materials design. Classical supercomputers traditionally have been the workhorses for these tasks; however, their computational cost grows exponentially with problem size and complexity. Quantum annealing presents an alternative paradigm. Unlike classical computations that rely on deterministic searches or probabilistic algorithms, quantum annealing exploits quantum fluctuations to navigate rugged energy landscapes efficiently, making it especially potent for identifying near-optimal solutions within complex, high-dimensional spaces.

This new study pivots away from exact optimization—which demands the absolute best solution and has proven stubbornly resistant to quantum acceleration—and instead embraces approximate optimization. This approach focuses on finding solutions that are near-optimal, within a tolerable margin typically defined as within 1% or less of the global optimum. Such an angle is not only more realistic for most practical applications but also amplifies the opportunity for quantum methods to deliver performance gains. Real-world problems, such as constructing stock portfolios or optimizing supply chains, often benefit more from quality solutions found quickly than from exact computations that may be infeasible or excessively time-consuming.

Central to the study was the deployment of a D-Wave Advantage quantum annealing processor housed at USC’s Information Sciences Institute. This specialized quantum device operates by encoding optimization problems into quantum spins and then evolving the system toward minimal energy configurations, which correlate to optimal or near-optimal problem solutions. Yet, like all current quantum devices, it faces challenges from noise and decoherence, which can obscure the subtle quantum effects necessary to gain computational edge. The research team countered this by applying an advanced error suppression technique known as quantum annealing correction (QAC), profoundly boosting the integrity of the quantum information being processed.

By introducing QAC, the researchers successfully created over 1,300 error-suppressed logical qubits. This unprecedented scale and error mitigation allowed the quantum annealer to outperform the most powerful classical optimization algorithm to date—parallel tempering with isoenergetic cluster moves (PT-ICM). PT-ICM is widely recognized for its efficiency in exploring complex energy landscapes typical of spin-glass models, making it the most formidable classical benchmark to beat. Surpassing PT-ICM illustrates not only theoretical significance but also practical promise, suggesting that quantum annealers can soon become indispensable tools in fields where optimization under uncertainty and noise tolerance is essential.

The team’s experimental methodology was rigorous and comprehensive, focusing on a well-defined family of two-dimensional spin-glass problems. Spin glasses represent a notoriously challenging class of disordered magnetic systems extensively studied in statistical physics for their complex, rugged energy landscapes. Their complexity makes them ideal testbeds for benchmarking advanced optimization algorithms, both classical and quantum. By focusing on these problems, the USC group ensured their findings had both fundamental scientific importance and broad applicability in various optimization contexts.

What sets this research apart is the innovative metric used to assess quantum advantage, termed “time-to-epsilon.” Unlike traditional metrics that measure raw computational time or solution accuracy in isolation, time-to-epsilon captures how long an algorithm takes to reliably find solutions within a specified epsilon margin of the optimal value. This nuanced performance measure aligns seamlessly with modern optimization requirements, where achieving near-optimal solutions quickly often outweighs exact but impractical perfection. Applying this metric revealed substantial scaling advantages for the quantum annealer as problem sizes increased, marking a credible demonstration of scaling quantum advantage—a phenomenon eagerly sought after in the quantum computing community.

Looking forward, the USC research group envisions extending their quantum advantage framework to tackle denser, higher-dimensional optimization problems that more closely reflect real-world complexities. Higher-dimensional instances naturally present exponential scaling complexities, historically thwarting classical approaches from delivering feasible solutions within reasonable timeframes. Achieving quantum advantage across these more intricate problem landscapes could revolutionize sectors ranging from logistics optimization and bioinformatics to artificial intelligence and quantum chemistry, essentially opening a new frontier in computational capability.

Furthermore, the research highlights the critical role of continued advances in quantum hardware fidelity and error suppression. Although quantum annealing has shown promising near-term applicability, its effectiveness is presently curtailed by noise and limited coherence times of existing qubit architectures. The quantum annealing correction technique represents a pivotal step in mitigating these issues, but ongoing innovations in quantum device engineering and error resilience will be vital to harnessing the full power of quantum optimization, potentially scaling quantum advantage beyond current horizons.

Daniel Lidar, the paper’s corresponding author and a professor at USC’s Viterbi School of Engineering and Dornsife College of Letters, Arts and Sciences, underscores the transformative potential of these findings. “This work not only validates quantum annealing as a competitive approach for approximate optimization but also paves the way for deploying quantum algorithms in practical applications where near-optimality is sufficient and computational speed is paramount,” he noted. His team’s research advances the narrative that quantum computing is gradually transitioning from theoretical promise to pragmatic, real-world impact.

The implications of this research resonate through industries where optimization problems abound. Financial modeling, where beating market indices rather than achieving absolute optimal portfolios suffices, exemplifies such potential use cases. Equally critical are logistics and supply chain management, where rapidly adapting schedules and distributions with near-optimal plans can save substantial resources and costs. By realigning quantum computing focus towards approximate, scalable solutions, this study significantly enriches the toolkit available to scientists and engineers facing immense computational challenges.

In conclusion, the USC study stands as a landmark demonstration of scaling quantum advantage in approximate optimization using quantum annealing. Its technical sophistication, combined with practical orientation, sets a new standard in quantum information science research. By advancing error suppression techniques, adopting realistic performance metrics, and benchmarking against the best classical algorithms, it showcases quantum computation’s growing maturity and readiness to address some of the most challenging problems society faces. As quantum annealers evolve and integrate into broader computational ecosystems, they promise not only to complement classical supercomputing but also to redefine the boundaries of what is computationally feasible.

Subject of Research: Not applicable
Article Title: Scaling Advantage in Approximate Optimization with Quantum Annealing
News Publication Date: 23-Apr-2025
Web References:

• DOI: 10.1103/PhysRevLett.134.160601
• USC News Article: Quantum computer outperforms supercomputers in approximate optimization

References:
Munoz-Bauza, H., & Lidar, D. (2025). Scaling Advantage in Approximate Optimization with Quantum Annealing. Physical Review Letters. DOI: 10.1103/PhysRevLett.134.160601

Keywords:
Quantum computing, Qubits, Algorithms, Annealing, Mathematical optimization