用模式匹配压缩时间序列，提速23倍还更准
Byte Pair Encoding for Efficient Time Series Forecasting
- 按常见模式合并数据点成令牌，动态压缩序列
- 平均预测性能提升40%，计算效率提高2314%
- 无需训练就能优化结果，适合各类时间序列任务
现有时间序列分词方法通常将固定数量样本编码为单一令牌，对简单模式（如长期恒定值）会产生过多令牌，导致显著计算开销。受字节对编码启发，我们提出首个面向时间序列分析的模式中心分词方案。基于频繁出现的模式词汇表，该方法将具有相似内在模式的样本合并为令牌，实现自适应压缩。结合有限模式集合与时间序列连续特性，进一步引入条件解码作为轻量级后处理优化方法，无需梯度计算且不增加计算开销。在近期时间序列基础模型上，基于模式的分词使预测性能平均提升40%，效率提升2314%；条件解码进一步将均方误差降低最多达48%。大量实验表明，该分词方法对多种时序模式具有自适应性，能泛化至未见数据，并生成捕捉统计矩、趋势等关键特性的有意义令牌表示。
原文摘要 · Abstract (English)
Existing time series tokenization methods predominantly encode a constant number of samples into individual tokens. This inflexible approach can generate excessive tokens for even simple patterns like extended constant values, resulting in substantial computational overhead. Inspired by the success of byte pair encoding, we propose the first pattern-centric tokenization scheme for time series analysis. Based on a discrete vocabulary of frequent motifs, our method merges samples with underlying patterns into tokens, compressing time series adaptively. Exploiting our finite set of motifs and the continuous properties of time series, we further introduce conditional decoding as a lightweight yet powerful post-hoc optimization method, which requires no gradient computation and adds no computational overhead. On recent time series foundation models, our motif-based tokenization improves forecasting performance by 40% and boosts efficiency by 2314% on average. Conditional decoding further reduces MSE by up to 48%. In an extensive analysis, we demonstrate the adaptiveness of our tokenization to diverse temporal patterns, its generalization to unseen data, and its meaningful token representations capturing distinct time series properties, including statistical moments and trends.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。