IndexNet: Timestamp and Variable-Aware Modeling for Time Series Forecasting

2025-09-28

AI Summary

Existing multivariate time series forecasting methods commonly neglect the contextual semantics embedded in timestamps and variable indices, thereby limiting their capacity to model periodic patterns and capture inter-variable heterogeneity. To address this, we propose IndexNet—a novel framework that systematically introduces plug-and-able timestamp embedding (TE) and channel embedding (CE) modules, jointly encoding temporal positions and variable identities within a lightweight MLP architecture. This design balances expressiveness and generality, significantly enhancing modeling capability for complex periodic structures and cross-variable heterogeneity. Evaluated on 12 real-world datasets, IndexNet achieves prediction accuracy comparable to state-of-the-art baselines. Ablation studies and visualization analyses confirm its effectiveness in capturing periodicity and distinguishing variables, while also improving model interpretability.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Temporal Planning
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Web crawling and indexing

Abstract

Multivariate time series forecasting (MTSF) plays a vital role in a wide range of real-world applications, such as weather prediction and traffic flow forecasting. Although recent advances have significantly improved the modeling of temporal dynamics and inter-variable dependencies, most existing methods overlook index-related descriptive information, such as timestamps and variable indices, which carry rich contextual semantics. To unlock the potential of such information and take advantage of the lightweight and powerful periodic capture ability of MLP-based architectures, we propose IndexNet, an MLP-based framework augmented with an Index Embedding (IE) module. The IE module consists of two key components: Timestamp Embedding (TE) and Channel Embedding (CE). Specifically, TE transforms timestamps into embedding vectors and injects them into the input sequence, thereby improving the model's ability to capture long-term complex periodic patterns. In parallel, CE assigns each variable a unique and trainable identity embedding based on its index, allowing the model to explicitly distinguish between heterogeneous variables and avoid homogenized predictions when input sequences seem close. Extensive experiments on 12 diverse real-world datasets demonstrate that IndexNet achieves comparable performance across mainstream baselines, validating the effectiveness of our temporally and variably aware design. Moreover, plug-and-play experiments and visualization analyses further reveal that IndexNet exhibits strong generality and interpretability, two aspects that remain underexplored in current MTSF research.

Problem

• Modeling timestamp and variable index information in time series
• Capturing long-term complex periodic patterns through timestamp embeddings
• Distinguishing heterogeneous variables to prevent homogenized predictions

Innovation

• IndexNet uses MLP with timestamp and variable embeddings
• TE captures long-term complex periodic patterns
• CE distinguishes variables to prevent homogenized predictions