PulseAugur
LIVE 07:37:49
research · [3 sources] ·
0
research

New PAMNet and PAMod frameworks improve time series forecasting with phase-amplitude modulation

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 3 sources

Researchers have introduced PAMod, a new framework designed to improve time series forecasting by addressing non-stationary data. PAMod models cyclical distribution shifts using phase-amplitude modulation in a normalized feature space. This approach adapts to changes in mean and variance, offering a more robust solution than previous methods like RevIN. Experiments show PAMod achieves state-of-the-art results with reduced computational cost and can be integrated into existing forecasting models. AI

Summary written by gemini-2.5-flash-lite from 3 sources. How we write summaries →

IMPACT Enhances time series forecasting accuracy and efficiency, potentially improving applications in finance, weather, and demand prediction.

RANK_REASON Academic paper introducing a new methodology for time series forecasting.

Read on arXiv cs.AI →

COVERAGE [3]

  1. arXiv cs.LG TIER_1 · Yingbo Zhou, Yutong Ye, Zhiwei Ling, Shuhao Li, Rui Qian, Jian Xiong, Li Sun, Dejing Dou ·

    PAMNet: Cycle-aware Phase-Amplitude Modulation Network for Multivariate Time Series Forecasting

    arXiv:2605.02938v1 Announce Type: new Abstract: Reliable periodic patterns serve as a fundamental basis for accurate multivariate time series forecasting. However, existing methods either implicitly extract periodicity through complex model architectures (e.g., Transformers) with…

  2. arXiv cs.LG TIER_1 · Yingbo Zhou, Yutong Ye, Shuhao Li, Rui Qian, Qiang Huang, Lemao Liu, Li Sun, Dejing Dou ·

    PAMod: Modeling Cyclical Shifts via Phase-Amplitude Modulation for Non-stationary Time Series Forecasting

    arXiv:2605.00466v1 Announce Type: new Abstract: Real-world time series forecasting faces the fundamental challenge of non-stationary statistical properties, including shifts in mean and variance over time. While reversible instance normalization (RevIN) has shown promise by stati…

  3. arXiv cs.AI TIER_1 · Dejing Dou ·

    PAMod: Modeling Cyclical Shifts via Phase-Amplitude Modulation for Non-stationary Time Series Forecasting

    Real-world time series forecasting faces the fundamental challenge of non-stationary statistical properties, including shifts in mean and variance over time. While reversible instance normalization (RevIN) has shown promise by stationarizing inputs and denormalizing outputs, it r…

RELATED ENTITIES

RELATED TOPICS