Auto-Configured Networks for Multi-Scale Multi-Output Time-Series Forecasting

2026-04-08

AI Summary

This work addresses the challenges of industrial time-series forecasting, where multi-source asynchronous signals, multiple output targets, and the explicit trade-off between prediction error and model complexity under deployment constraints hinder existing methods from jointly optimizing preprocessing, architecture, and hyperparameters within limited budgets. To tackle this, the authors propose an automated configuration framework featuring a multi-scale dual-branch convolutional neural network (MS-BCNN) that jointly captures local fluctuations and long-term trends. They unify alignment strategies, network architectures, and hyperparameters into a hierarchical conditional mixed search space and introduce a Player-based hybrid multi-objective evolutionary algorithm (PHMOEA) to efficiently approximate the Pareto front between error and complexity. Experiments on both synthetic and real-world sintering datasets demonstrate superior performance over state-of-the-art baselines and support flexible deployment scenarios.

Abstract

Industrial forecasting often involves multi-source asynchronous signals and multi-output targets, while deployment requires explicit trade-offs between prediction error and model complexity. Current practices typically fix alignment strategies or network designs, making it difficult to systematically co-design preprocessing, architecture, and hyperparameters in budget-limited training-based evaluations. To address this issue, we propose an auto-configuration framework that outputs a deployable Pareto set of forecasting models balancing error and complexity. At the model level, a Multi-Scale Bi-Branch Convolutional Neural Network (MS--BCNN) is developed, where short- and long-kernel branches capture local fluctuations and long-term trends, respectively, for multi-output regression. At the search level, we unify alignment operators, architectural choices, and training hyperparameters into a hierarchical-conditional mixed configuration space, and apply Player-based Hybrid Multi-Objective Evolutionary Algorithm (PHMOEA) to approximate the error--complexity Pareto frontier within a limited computational budget. Experiments on hierarchical synthetic benchmarks and a real-world sintering dataset demonstrate that our framework outperforms competitive baselines under the same budget and offers flexible deployment choices.

Problem

• multi-output forecasting
• model complexity
• error-complexity trade-off
• asynchronous time series
• Pareto optimization

Innovation

• Multi-Scale Bi-Branch CNN
• Auto-Configuration
• Pareto-Optimal Forecasting
• Hierarchical-Conditional Search Space
• Multi-Objective Evolutionary Algorithm