Channel Normalization for Time Series Channel Identification

TL;DR

This paper introduces Channel Normalization (CN) and its variants to improve channel identifiability in time series models, leading to enhanced performance and adaptability across different datasets and models.

Contribution

It proposes a novel normalization strategy, Channel Normalization, with adaptive and prototypical extensions, to enhance channel identifiability in time series modeling.

Findings

Significant performance improvements in various TS models.

Enhanced adaptability with Adaptive CN (ACN).

Effective handling of datasets with unknown or varying channels.

Abstract

Channel identifiability (CID) refers to the ability to distinguish between individual channels in time series (TS) modeling. The absence of CID often results in producing identical outputs for identical inputs, disregarding channel-specific characteristics. In this paper, we highlight the importance of CID and propose Channel Normalization (CN), a simple yet effective normalization strategy that enhances CID by assigning distinct affine transformation parameters to each channel. We further extend CN in two ways: 1) Adaptive CN (ACN) dynamically adjusts parameters based on the input TS, improving adaptability in TS models, and 2) Prototypical CN (PCN) introduces a set of learnable prototypes instead of per-channel parameters, enabling applicability to datasets with unknown or varying number of channels and facilitating use in TS foundation models. We demonstrate the effectiveness of CN and its variants by applying them to various TS models, achieving significant performance gains for both non-CID and CID models. In addition, we analyze the success of our approach from an information theory perspective. Code is available at https://github.com/seunghan96/CN.