Quantum models improve time series classification with modular circuits
Towards Scaling Quantum Fine-Tuning of Foundational Time Series Models for Classification
Emerging Technologies
Summary
Classifying time-based data like power grid events is challenging, and people use special models for this. The authors explored how quantum components can improve these models by fine-tuning them with quantum circuits that process the model’s data summaries. They found that adding small extra circuits called wings, each handling new input information, improved accuracy beyond classical methods and larger quantum circuits without new data. This shows that quantum models get better when they process more types of information, not just bigger circuits with more parameters.
time seriesfoundation modelsquantum circuitsfine-tuningpower grid classificationembeddingsbalanced accuracyqubitsmodular architecturesmultilayer perceptron
Authors
Sang Hyub Kim, Julien Baglio, Rajiv Krishnakumar, Chi Chen, Oliver Knitter, Jonathan Mei, Claudio Girotto, Masako Yamada, Frederik F. Flöther, Martin Roetteler
Abstract
Time-series foundation models produce rich embeddings, but whether quantum models can exploit them, and how far hybrid classical-quantum architectures scale, remains unclear. We address this by fine-tuning Chronos for power-grid event classification (PSML-5) with a quantum head on the model's embeddings. Grouping embeddings by physical sensor type before summarization already surpasses the best published baseline built for this benchmark, and with finer-grained features the quantum head outperforms a larger classical multilayer perceptron on identical inputs by 1.7-2.0 percentage points of balanced accuracy. Yet the gains saturate: past a point, feeding more information to the same fixed-width register yields no improvement. We show the bottleneck is neither the supply of information nor circuit expressiveness, but the bandwidth of the data intake. To overcome this limitation, we introduce the wing module, a self-contained few-qubit circuit that feeds additional information into the core circuit through a sparse, one-way coupling. Under a preregistered four-seed protocol, we attach wings to a fixed 12-qubit core with fixed features. Balanced accuracy increases with each added wing, from 83.6% with no wings (13 qubits, including a post-selection qubit) to 85.2% with two (19 qubits). Ablations establish that a circuit enlarged without new information gains nothing, while a wing fed information from the wrong sample harms accuracy. These results reframe scaling for quantum fine-tuning: added qubits help when they carry added inputs, not merely more parameters. Wings offer a modular and stable route to widening that bandwidth.