Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Hybrid Quantum Temporal Convolutional Networks

Junghoon Justin Park, Maria Pak, Sebin Lee, Samuel Yen-Chi Chen, Shinjae Yoo, Huan-Hsin Tseng, Jiook Cha

TL;DR

The Hybrid Quantum Temporal Convolutional Network (HQTCN), which combines classical temporal windowing with a quantum convolutional neural network core, is introduced, which establishes HQTCN as a parameter-efficient approach for multivariate time-series analysis.

Abstract

Quantum machine learning models for sequential data face scalability challenges with complex multivariate signals. We introduce the Hybrid Quantum Temporal Convolutional Network (HQTCN), which combines classical temporal windowing with a quantum convolutional neural network core. By applying a shared quantum circuit across temporal windows, HQTCN captures long-range dependencies while achieving significant parameter reduction. Evaluated on synthetic NARMA sequences and high-dimensional EEG time-series, HQTCN performs competitively with classical baselines on univariate data and outperforms all baselines on multivariate tasks. The model demonstrates particular strength under data-limited conditions, maintaining high performance with substantially fewer parameters than conventional approaches. These results establish HQTCN as a parameter-efficient approach for multivariate time-series analysis.

Hybrid Quantum Temporal Convolutional Networks

TL;DR

The Hybrid Quantum Temporal Convolutional Network (HQTCN), which combines classical temporal windowing with a quantum convolutional neural network core, is introduced, which establishes HQTCN as a parameter-efficient approach for multivariate time-series analysis.

Abstract

Quantum machine learning models for sequential data face scalability challenges with complex multivariate signals. We introduce the Hybrid Quantum Temporal Convolutional Network (HQTCN), which combines classical temporal windowing with a quantum convolutional neural network core. By applying a shared quantum circuit across temporal windows, HQTCN captures long-range dependencies while achieving significant parameter reduction. Evaluated on synthetic NARMA sequences and high-dimensional EEG time-series, HQTCN performs competitively with classical baselines on univariate data and outperforms all baselines on multivariate tasks. The model demonstrates particular strength under data-limited conditions, maintaining high performance with substantially fewer parameters than conventional approaches. These results establish HQTCN as a parameter-efficient approach for multivariate time-series analysis.
Paper Structure (20 sections, 10 equations, 2 figures, 3 tables)

This paper contains 20 sections, 10 equations, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Recurrent Quantum Models
  4. Classical Temporal Convolutional Networks (TCNs)
  5. Quantum Convolutional Neural Network (QCNN)
  6. Hybrid Quantum Temporal Convolutional Networks
  7. Dilated Temporal Windowing
  8. The Linear Embedding Interface
  9. Shared Quantum Convolutional Core
  10. Output Aggregation
  11. Experiment
  12. Data
  13. NARMA
  14. PhysioNet EEG
  15. Experimental Setup
  16. ...and 5 more sections

Figures (2)

  • Figure 1: Model Architectures. (a) A standalone QCNN with hierarchical quantum convolutional and pooling layers. (b) The proposed HQTCN, which applies a shared QCNN circuit over a temporal sliding window before aggregating the outputs.
  • Figure 2: NARMA Time-Series Prediction Results. Comparison of HQTCN's predictions (red) against the ground truth (gray) and other models on the NARMA test set. The zoomed-in view highlights the test-set performance.