TIGER: Time-Series Classification with In-Context-Learning Gated Ensemble of Representations
作者: Johann Faouzi
分类: cs.LG
发布日期: 2026-10-05
💡 一句话要点
提出TIGER以解决时间序列分类中的适应性不足问题
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 时间序列分类 自适应学习 集成算法 机器学习 特征提取
📋 核心要点
- 现有的时间序列分类方法,如HIVE-COTE~2.0,通常依赖于固定的非自适应规则,导致在不同数据集上的表现不佳。
- TIGER通过使用三种通用分类器对四种不同的表示进行处理,并采用自适应的元分类规则,提升了分类的灵活性和准确性。
- 在142个数据集的基准测试中,TIGER在平均准确率、平衡准确率和F1-score上均优于六种对比算法,显示出显著的性能提升。
📝 摘要(中文)
本文提出了一种新的时间序列分类方法TIGER(基于上下文学习的表示集成),旨在克服现有集成算法在时间序列分类中的局限性。与现有的HIVE-COTE~2.0等方法不同,TIGER采用了三种通用分类器(岭回归、额外树和朴素贝叶斯)对四种不同表示进行处理,并通过自适应的元分类规则生成最终预测。实验结果表明,TIGER在142个数据集的基准测试中表现优异,超越了包括HIVE-COTE~2.0在内的六种对比算法,且其单一超参数在全评估基准中具有良好的泛化能力。
🔬 方法详解
问题定义:本文旨在解决时间序列分类中现有集成算法的适应性不足问题,尤其是HIVE-COTE~2.0等方法在不同数据集上表现不佳的挑战。
核心思路:TIGER采用了三种通用分类器(岭回归、额外树和朴素贝叶斯),对四种不同的表示进行处理,生成12个基础学习器的预测,并通过自适应的元分类规则进行最终预测。
技术框架:整体架构包括四个表示的提取、三种分类器的应用、基础学习器预测的堆叠,以及自适应元分类规则的选择,形成一个完整的分类流程。
关键创新:TIGER的主要创新在于其自适应元分类规则,能够根据每个数据集的训练样本数量选择加权多数投票或TabICLv2作为最终预测方法,显著提升了分类性能。
关键设计:TIGER的设计中包含一个单一超参数,通过在20个数据集的开发子集上进行调优,展示了其在全评估基准中的良好泛化能力。
🖼️ 关键图片
📊 实验亮点
在142个数据集的基准测试中,TIGER在平均准确率、平衡准确率和F1-score上均优于六种对比算法,包括HIVE-COTE~2.0,显示出显著的性能提升,且其自适应规则在单独使用的两种元分类方法中也表现出更好的效果。
🎯 应用场景
TIGER在时间序列分类领域具有广泛的应用潜力,适用于金融市场分析、医疗监测、环境监测等多个领域。其自适应的分类机制能够根据不同数据集的特性进行调整,提升分类的准确性和可靠性,未来可能推动相关领域的智能决策和预测技术的发展。
📄 摘要(原文)
A representation family is a distinct way of extracting features from time series. Ensemble algorithms that combine several representation families remain the most accurate approach to time series classification. Current state-of-the-art ensembles, most notably HIVE-COTE~2.0, pair a bespoke classification algorithm with each representation family and combine their predictions using a fixed, non-adaptive rule. We present TIGER (Time-series classification with In-context-learning Gated Ensemble of Representations), which instead applies the same small portfolio of three general-purpose classifiers (Ridge, Extra Trees, and Naive Bayes) to four representations from four distinct families, stacking the resulting twelve base learners' predictions into a meta-feature matrix. The final prediction is produced by an adaptive meta-classification rule that chooses, independently for each data set, between a weighted hard majority vote and TabICLv2, a pretrained tabular foundation model used in-context as a meta-classifier, based on the mean number of training samples available per class. On a 142-data-set benchmark drawn from the UCR time series classification archive, TIGER obtains the best mean accuracy, balanced accuracy, and F1-score among six compared algorithms, including HIVE-COTE~2.0, and significantly outperforms each of the other five individually. TIGER's adaptive rule also meaningfully outperforms either of its two constituent meta-classification methods used alone, and its single hyperparameter, tuned using only a twenty-data-set development subset, is shown to generalize to the full evaluation benchmark. We further characterize TIGER's design through an extensive set of ablation experiments and report the design alternatives that we investigated and ultimately discarded.