Metraipolis › archive › dc_5a99bc0db0be

Koopman Generative Operators for Efficient Probabilistic Time-Series Forecasting — arXiv:2611.04760

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.04760
License (as archived): CC0-1.0
Archived: 2026-11-04T12:40:56.654Z
SHA-256 of the archived text: 41bec1e30006d484ef503a49ac249cc9d16de51ba9ec489b5d3dc2362108defb
Stable URL of this page: https://metraipolis.com/documento/dc_5a99bc0db0be

How to cite

Koopman Generative Operators for Efficient Probabilistic Time-Series Forecasting — arXiv:2611.04760 — arXiv (en), https://arxiv.org/abs/2611.04760 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5a99bc0db0be (Metraipolis archive dc_5a99bc0db0be, sha256 41bec1e30006d484ef503a49ac249cc9d16de51ba9ec489b5d3dc2362108defb)

BibTeX

@misc{dc5a99bc0db0be,
  title = {Koopman Generative Operators for Efficient Probabilistic Time-Series Forecasting — arXiv:2611.04760},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04760},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5a99bc0db0be; sha256 41bec1e30006d484ef503a49ac249cc9d16de51ba9ec489b5d3dc2362108defb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04760. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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 25$\times$ 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.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.04760 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-04T12:29:47.306Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).