Sep 2026· Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence· 0 citations· 44 references
TL;DR
This paper re-examines frequency-domain modeling from the perspective of spectral non-identifiability and proposes a Spectral Memory Learner with Dual Retrieval (SMLDR) for time series forecasting, which represents periodic components as reusable spectral prototypes stored in learnable memory.
Abstract
Time series forecasting aims to predict future values using historical observations, which is crucial for many practical applications with complex temporal dynamics. Recent frequency-domain forecasting methods have utilized spectral representations to model periodicity, but they usually rely on an implicit assumption: periodic components and aperiodic residuals are orthogonal in the frequency domain. However, in practice, this assumption often does not hold due to issues such as spectral leakage, superposition of multiple periods, and the gradual periodic changes. This paper re-examines frequency-domain modeling from the perspective of spectral non-identifiability and proposes a Spectral Memory Learner with Dual Retrieval (SMLDR) for time series forecasting. Instead of modeling periodicity in the frequency dimension, SMLDR represents periodic components as reusable spectral prototypes stored in learnable memory. The dual retrieval mechanism extracts historical periodic components and predicts them in the frequency domain, while a lightweight time-domain branch is used to model aperiodic dynamics to predict future residuals.Extensive experiments on multiple time series forecasting benchmarks show that SMLDR consistently achieves excellent performance, and the learnable memory can capture complex periodicity.
To address the sequential and evolving nature of time series, the Online Time Series Forecasting (OTSF) task has been extensively studied in multiple domains. Existing research focuses on adapting to non-stationary environments by employing memory buffer-based retrieval strategies. However, we observe that such framewo...
Seungyoon Choi, Hyunchul C. Kim, Jae-Gil Lee et al.· 1 citation
This work derives a parameterizable evolutionary spectra formulation, recasting non-stationary random process modeling as learning an evolving representation modulated by random variables, and reduces the complexity of the estimated spectra from O(NM) to O(NK), where K<<M/2.
Wei-Wei Ye, Ren-He Jiang, Hang-Chen Liu et al.· 0 citations
Irregular time series forecasting is crucial in many domains, such as healthcare and meteorological observation. However, due to the inherent characteristics of irregular time series, including sparse observations and non-uniform sampling, accurately predicting future dynamics remains challenging. In light of these two...
PAC-LLM is proposed, a phase-space-aware adaptive fusion framework for long-term chaotic time series forecasting powered by LLMs that leverages learned phase-space features and textual information to fully enable LLM's time series forecasting capacity.
This work introduces a novel sampling-conditioned temporal manifold attention mechanism that maps real timestamps into a learnable temporal manifold feature space and injects temporal manifold biases into attention layers, enabling the model to capture both irregular time intervals and local sampling structures.
Lin-Feng Wang, Rui-Tong Zhang, Kai Zhao et al.· 0 citations
DecoVAE is proposed, a lightweight interpretable trend-seasonal VAE framework that explicitly decomposes time series into trend and seasonal components by applying domain-specific inductive biases.
A. Marusov, D. Anikin, A. Zaytsev· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.