MSTAR: Multi-Scale Backbone Architecture Search for Timeseries Classification

TL;DR

This paper introduces MSTAR, a neural architecture search framework for time series classification that optimizes multi-scale receptive fields, achieving state-of-the-art results across diverse datasets.

Contribution

It proposes a novel multi-scale search space and NAS framework that automatically discovers optimal scales for time series data, addressing limitations of manual design and scalability.

Findings

Achieves state-of-the-art performance on four datasets across different domains.

Introduces over ten fine-tuned models for each dataset.

Demonstrates the effectiveness of multi-scale NAS in time series classification.

Abstract

Most of the previous approaches to Time Series Classification (TSC) highlight the significance of receptive fields and frequencies while overlooking the time resolution. Hence, unavoidably suffered from scalability issues as they integrated an extensive range of receptive fields into classification models. Other methods, while having a better adaptation for large datasets, require manual design and yet not being able to reach the optimal architecture due to the uniqueness of each dataset. We overcome these challenges by proposing a novel multi-scale search space and a framework for Neural architecture search (NAS), which addresses both the problem of frequency and time resolution, discovering the suitable scale for a specific dataset. We further show that our model can serve as a backbone to employ a powerful Transformer module with both untrained and pre-trained weights. Our search space reaches the state-of-the-art performance on four datasets on four different domains while introducing more than ten highly fine-tuned models for each data.