TimeMIL: Advancing Multivariate Time Series Classification via a Time-aware Multiple Instance Learning

TL;DR

TimeMIL introduces a weakly supervised, time-aware multiple instance learning framework using transformers and wavelet positional tokens to improve multivariate time series classification and pattern localization.

Contribution

It is the first to reformulate MTSC as a weakly supervised problem with a novel time-aware MIL pooling and transformer-based architecture.

Findings

Outperformed 26 recent state-of-the-art methods.

Effectively localized patterns in time series data.

Demonstrated robustness in pattern detection.

Abstract

Deep neural networks, including transformers and convolutional neural networks, have significantly improved multivariate time series classification (MTSC). However, these methods often rely on supervised learning, which does not fully account for the sparsity and locality of patterns in time series data (e.g., diseases-related anomalous points in ECG). To address this challenge, we formally reformulate MTSC as a weakly supervised problem, introducing a novel multiple-instance learning (MIL) framework for better localization of patterns of interest and modeling time dependencies within time series. Our novel approach, TimeMIL, formulates the temporal correlation and ordering within a time-aware MIL pooling, leveraging a tokenized transformer with a specialized learnable wavelet positional token. The proposed method surpassed 26 recent state-of-the-art methods, underscoring the effectiveness of the weakly supervised TimeMIL in MTSC. The code will be available at https://github.com/xiwenc1/TimeMIL.