Quantum neural networks improve time series and learning tasks
Learning to Program Adaptive Non-Local Observables for Machine Learning
Machine Learning
Summary
Quantum neural networks usually use fixed ways to measure their outputs, which can limit how well they work. The authors created a method that lets the network change how it measures information based on each input, making it more flexible. They tested this on predicting complex time-based data and in reinforcement learning tasks, finding that their approach performed better than previous methods. This shows that adapting measurements in quantum networks can help them learn and predict more accurately.
What this means in practice
- •For time-series analysts: Enhance forecasting accuracy for complex multivariate time-series using input-adaptive quantum neural networks.
- •For reinforcement learning engineers: Improve decision-making in reinforcement learning tasks by adapting quantum model measurements dynamically based on input.
Authors
Yu-Ting Lee, Samuel Yen-Chi Chen, Huan-Hsin Tseng
Abstract
Quantum neural networks (QNNs) are typically built from variational quantum circuits (VQCs), which are limited by local measurements. Adaptive non-local observables (ANO) address this by jointly optimizing circuit parameters and multi-qubit measurements. However, existing ANO-based VQCs learn only a single static observable that remains invariant across all inputs. We propose QFWP-ANO, a novel architecture which employs a classical hypernetwork to dynamically program VQC parameters and/or non-local observables conditioned on each input. On multivariate time-series forecasting across four ETT datasets, QFWP-ANO achieves the lowest MSE in 16 of 20 settings and second-lowest in the remaining four, surpassing ANO-based and other strong baselines. On reinforcement learning tasks, QFWP-ANO consistently surpasses ANO-VQCs. Our results establish input-conditioned ANO as an effective approach for enhancing QNNs.