Multivariate Time Series Forecasting with Adaptive Non-Local Observables
2026-07-27 • Artificial Intelligence
Artificial IntelligenceMachine Learning
AI summaryⓘ
The authors developed a new method, called MTSF-ANO, to predict multiple time series data more accurately using quantum neural networks. Unlike previous quantum models that use fixed local measurements, their approach uses flexible, non-local measurements which help capture more complex patterns. Tested on four datasets, MTSF-ANO often performed better than existing methods, sometimes improving accuracy by up to 20%. Their experiments also showed the importance of their quantum circuit design and the use of adaptive measurements for better results.
Multivariate time series forecastingQuantum neural networksVariational quantum circuitsAdaptive non-local observablesMean squared errorETT datasetHybrid quantum-classical modelsQuantum circuit designTime series predictionAblation study
Authors
Yu-Ting Lee, Huan-Hsin Tseng, Samuel Yen-Chi Chen
Abstract
Multivariate time series forecasting (MTSF) predicts future values of multiple variables from historical data. While quantum neural networks have been increasingly applied to this task, they typically rely on fixed local measurements, which restrict their expressivity. We propose MTSF-ANO, a simple hybrid model for MTSF that integrates variational quantum circuits with adaptive non-local observables (ANO). On the four ETT datasets, MTSF-ANO ranks first or second in MSE in 17 of 20 settings, improving over the strongest baseline by up to 20% on ETTh1, and outperforms or matches its fixed local observable counterpart across all settings. Our ablations show how the quantum circuit design and ANO non-locality affect performance. These results suggest that ANO is a promising direction for quantum time series forecasting.