跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 19 天前AI 評分35

Koopman 生成算子:高效概率時間序列預測

Koopman Generative Operators for Efficient Probabilistic Time-Series Forecasting

AI 導讀

Koopman Generative Operator(KGO)結合結構化潛在動態、單步 flow matching 和自適應不確定性門控,用於概率時間序列預測。KGO 在 17 項 ProbTS 基準中,CRPS 有 13 項、NMAE 有 16 項達到最先進水平。與迭代生成模型相比,KGO 的推理時間至少減少 25 倍。

正文

Published on Nov 3

Authors:

,

,

,

Abstract

KGO unifies structured latent dynamics with single-step flow matching and adaptive uncertainty gating to achieve accurate, efficient probabilistic time-series forecasting.

Probabilistic time-series forecasting requires models that simultaneously capture structured temporal dynamics, expressive uncertainty, and efficient inference, yet existing approaches fall short of this goal: latent dynamical models impose structure but rely on restrictive generation mechanisms, while modern generative methods such as diffusion and flow matching achieve flexibility at the cost of iterative and computationally expensive sampling. We introduce the Koopman Generative Operator (KGO), a probabilistic forecasting method that conceptualizes prediction as the evolution of structured uncertainty. KGO integrates three core components: (1) Koopman Patch Embedding (KoPE) for temporally consistent latent trajectory extrapolation; (2) Koopman Flow Matching (KoFM), which enables fast, single-step generation through a closed-form matrix exponential in a Koopman latent space, bypassing iterative sampling; and (3) an Adaptive Uncertainty Gate (AUG), which provides probabilistic forecasts by adapting uncertainty per-variable and per-horizon, corresponding to a learned decomposition of aleatoric uncertainty without manual tuning. By unifying these components, KGO achieves state-of-the-art accuracy on 13/17 ProbTS benchmarks in CRPS and 16/17 in NMAE. Further, by eliminating iterative sampling, KGO delivers at least a 25times reduction in inference time compared to iterative generative models. These results position KGO as a practical and principled framework that advances the accuracy-efficiency trade-off in probabilistic time-series forecasting, combining leading accuracy on the evaluated benchmarks with fast, single-step generation.

View arXiv page View PDF GitHub Add to collection

Get this paper in your agent:

hf papers read 2611.04760

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2611.04760 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2611.04760 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2611.04760 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co