← Back to all news
arXiv AI By Zhixuan Li, Naipeng Chen, Seonghwa Choi, Sanghoon Lee, Weisi Lin

BEAT: Balanced Frequency Adaptive Tuning for Long-Term Time-Series Forecasting

Read the original on arXiv AI →

BEAT (Balanced Frequency Adaptive Tuning) is a framework for long‑term time‑series forecasting that uses frequency‑specific error monitoring and adaptive gradient modulation. It introduces a Frequency‑Specific Monitor to compare predicted and target wavelet coefficients in a normalized space, and a Dynamical Gradient Balancer that assigns larger gradient weights to components with higher relative errors. Experiments on seven real‑world datasets demonstrate that BEAT achieves competitive performance against state‑of‑the‑art forecasting methods.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Machine Learning
Jun 2

FAiT: Frequency-Aware Inverted Transformer for Multivariate Time Series Forecasting

arXiv:2606. 01306v1 Announce Type: new Abstract: While Transformer-based architectures have established themselves as a dominant paradigm in Multivariate Time Series Forecasting (MTSF), their core self-attention mechanism inherently functions as a low-pass filter, systematically smoothing out high-frequency signals vital for sharp local changes.

By Peng He, Yao Liu, Yanglei Gan, Run Lin, Yuxiang Cai, Qiao Liu
llmsbenchmarkssafety
More like this →
arXiv Machine Learning
Aug 27

Frequency-aware forecasting for short-term typhoon gust prediction

The paper introduces WDANet, a frequency‑aware forecasting framework that uses stationary wavelet decomposition, FiLM, and a dual‑branch encoder‑decoder to separately model trend and fluctuation components in typhoon gust prediction. Applied to offshore Western Pacific wind data, WDANet outperforms ECMWF‑HRES for short lead times, achieving higher accuracy within the first 6 hours and better RMSE/MAE during extreme wind events. The study suggests WDANet could improve offshore wind power operations, disaster warnings, and risk mitigation.

By Xuefei Wang, Tingyi Liu, Heng Zhang, Shengjun Zhang
More like this →
arXiv Machine Learning
4d ago

GateTS: Versatile and Efficient Forecasting via Attention-Inspired routed Mixture-of-Experts

GateTS introduces a streamlined univariate forecasting model that merges sparse Mixture-of-Experts computation with an attention-inspired gating mechanism, eliminating the need for auxiliary load‑balancing losses. The architecture achieves superior predictive accuracy across both short‑ and long‑term horizons, including intermittent signals, while using far fewer parameters than leading transformer models such as PatchTST. Empirical tests on diverse datasets show GateTS is computationally more efficient than LSTM, enabling cost‑effective inference for real‑world time‑series applications.

By Kyrylo Yemets, Mykola Lukashchuk, Ivan Izonin
llmsbenchmarks
More like this →