轻量级时间序列模型FLAME，支持精准的确定性和概率预测。
FLAME: Flow Enhanced Legendre Memory Models for General Time Series Forecasting
- 用勒让德记忆捕捉数据内在规律，提升长程推理效率。
- 在多个基准上实现零样本最优性能，尤其在概率预测中表现突出。
- 适合需要高效、鲁棒预测的工业场景或资源受限环境。
本文提出FLAME，一类极轻量且强大的时间序列基础模型，通过生成式概率建模同时支持确定性和概率预测，兼顾效率与鲁棒性。FLAME采用勒让德记忆（Legendre Memory）实现强泛化能力，通过在编码和解码阶段引入平移勒让德（LegT）和缩放勒让德（LegS）变体，有效捕获数据中的归纳偏置，并实现高效的长程推理。为在保持高效的同时提升概率预测精度，FLAME采用基于归一化流（Normalization Flow）的预测头，以生成方式建模预测区间内任意复杂分布。在TSFM-Bench和ProbTS等权威基准上的全面实验表明，FLAME在确定性和概率预测任务中均表现出一致的领先零样本性能。
原文摘要 · Abstract (English)
In this work, we introduce FLAME, a family of extremely lightweight and capable Time Series Foundation Models, which support both deterministic and probabilistic forecasting via generative probabilistic modeling, thus ensuring both efficiency and robustness. FLAME utilizes the Legendre Memory for strong generalization capabilities. Through adapting variants of Legendre Memory, i.e., translated Legendre (LegT) and scaled Legendre (LegS), in the Encoding and Decoding phases, FLAME can effectively capture the inherent inductive bias within data and make efficient long-range inferences. To enhance the accuracy of probabilistic forecasting while keeping efficient, FLAME adopts a Normalization Flow based forecasting head, which can model the arbitrarily intricate distributions over the forecasting horizon in a generative manner. Comprehensive experiments on well-recognized benchmarks, including TSFM-Bench and ProbTS, demonstrate the consistent state-of-the-art zero-shot performance of FLAME on both deterministic and probabilistic forecasting tasks.
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。