IConv: Focusing on Local Variation with Channel Independent Convolution for Multivariate Time Series Forecasting

2025-09-25

AI Summary

Non-stationarity in multivariate time series—manifesting as local variations (e.g., seasonality, residuals)—hampers accurate modeling of channel-specific dynamics. To address this, we propose IConv: a novel architecture featuring channel-independent convolutional layers with large kernels to capture fine-grained local patterns per channel; a decoupled cross-channel interaction mechanism that preserves channel specificity while modeling global dependencies; and a hybrid design integrating MLPs and CNNs to jointly model long-term trends and short-term fluctuations. This design reduces computational overhead while enhancing representational capacity. Extensive experiments on multiple benchmark datasets demonstrate that IConv consistently outperforms state-of-the-art methods in forecasting accuracy—particularly excelling in capturing complex local patterns—establishing new performance benchmarks for multivariate time series forecasting.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Search and Optimization: Non-convex Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Real-world time-series data often exhibit non-stationarity, including changing trends, irregular seasonality, and residuals. In terms of changing trends, recently proposed multi-layer perceptron (MLP)-based models have shown excellent performance owing to their computational efficiency and ability to capture long-term dependency. However, the linear nature of MLP architectures poses limitations when applied to channels with diverse distributions, resulting in local variations such as seasonal patterns and residual components being ignored. However, convolutional neural networks (CNNs) can effectively incorporate these variations. To resolve the limitations of MLP, we propose combining them with CNNs. The overall trend is modeled using an MLP to consider long-term dependencies. The CNN uses diverse kernels to model fine-grained local patterns in conjunction with MLP trend predictions. To focus on modeling local variation, we propose IConv, a novel convolutional architecture that processes the temporal dependency channel independently and considers the inter-channel relationship through distinct layers. Independent channel processing enables the modeling of diverse local temporal dependencies and the adoption of a large kernel size. Distinct inter-channel considerations reduce computational cost. The proposed model is evaluated through extensive experiments on time-series datasets. The results reveal the superiority of the proposed method for multivariate time-series forecasting.

Problem

• Addressing limitations of MLP models in handling local variations in multivariate time series
• Modeling diverse seasonal patterns and residual components ignored by linear MLP architectures
• Combining MLP for long-term trends with CNN for fine-grained local pattern capture

Innovation

• Combines MLP and CNN for trend and local patterns
• Uses channel-independent convolution for temporal dependencies
• Employs distinct layers for inter-channel relationship modeling