paper-with-me

홈 › Papers

LightTS: Lightweight Time Series Classification with Adaptive Ensemble Distillation -- Extended Version

2023-02-24 · David Campos, Miao Zhang, Bin Yang, Tung Kieu, Chenjuan Guo, Christian S. Jensen

Due to the sweeping digitalization of processes, increasingly vast amounts of time series data are being produced. Accurate classification of such time series facilitates decision making in multiple domains. State-of-the-art classification accuracy is often achieved by ensemble learning where results are synthesized from multiple base models. This characteristic implies that ensemble learning needs substantial computing resources, preventing their use in resource-limited environments, such as in edge devices. To extend the applicability of ensemble learning, we propose the LightTS framework that compresses large ensembles into lightweight models while ensuring competitive accuracy. First, we propose adaptive ensemble distillation that assigns adaptive weights to different base models such that their varying classification capabilities contribute purposefully to the training of the lightweight model. Second, we propose means of identifying Pareto optimal settings w.r.t. model accuracy and model size, thus enabling users with a space budget to select the most accurate lightweight model. We report on experiments using 128 real-world time series sets and different types of base models that justify key decisions in the design of LightTS and provide evidence that LightTS is able to outperform competitors.

📄 PDF Abstract BibTeX arXiv:2302.12721

Code (1)

d-gcc/distiller 공식 구현 pytorch

Tasks

ClassificationDecision MakingEnsemble LearningTime SeriesTime Series AnalysisTime Series Classification

Methods 이 논문이 사용한 방법론

BASE 설명 없음

Similar Papers 제목 키워드 기반

Less Is More: Fast Multivariate Time Series Forecasting with Light Sampling-oriented MLP Structures

2022-07-04 · Tianping Zhang, Yizhuo Zhang, Wei Cao, Jiang Bian 외

Multivariate time series forecasting has seen widely ranging applications in various domains, including finance, traffic, energy, and healthcare. To capture the sophisticated temporal patterns, plenty of research studies…

Multivariate Time Series ForecastingTime SeriesTime Series AnalysisTime Series Forecasting

Temporal Patch Shuffle (TPS): Leveraging Patch-Level Shuffling to Boost Generalization and Robustness in Time Series Forecasting

2026-04-10 · Jafar Bakhshaliyev, Johannes Burchert, Niels Landwehr, Lars Schmidt-Thieme arxiv

Data augmentation is a crucial technique for improving model generalization and robustness, particularly in deep learning models where training data is limited. Although many augmentation methods have been developed for …

Time Series ClassificationTime Series ForecastingData Augmentation

Adaptive Law-Based Transformation (ALT): A Lightweight Feature Representation for Time Series Classification

2025-01-16 · Marcell T. Kurbucz, Balázs Hajós, Balázs P. Halmos, Vince Á. Molnár 외

Time series classification (TSC) is fundamental in numerous domains, including finance, healthcare, and environmental monitoring. However, traditional TSC methods often struggle with the inherent complexity and variabili…

Time SeriesTime Series Classification

ALT: A Python Package for Lightweight Feature Representation in Time Series Classification

2025-04-17 · Balázs P. Halmos, Balázs Hajós, Vince Á. Molnár, Marcell T. Kurbucz 외

We introduce ALT, an open-source Python package created for efficient and accurate time series classification (TSC). The package implements the adaptive law-based transformation (ALT) algorithm, which transforms raw time…

BenchmarkingTime SeriesTime Series Classification

Impact of Deep Learning Libraries on Online Adaptive Lightweight Time Series Anomaly Detection

2023-04-30 · Ming-Chang Lee, Jia-Chun Lin

Providing online adaptive lightweight time series anomaly detection without human intervention and domain knowledge is highly valuable. Several such anomaly detection approaches have been introduced in the past years, bu…

Anomaly DetectionDeep LearningTime SeriesTime Series Anomaly Detection