Wavelet Mixture of Experts for Time Series Forecasting

• URL: http://arxiv.org/abs/2508.08825v1
• Date: Tue, 12 Aug 2025 10:32:51 GMT
• Title: Wavelet Mixture of Experts for Time Series Forecasting
• Abstract summary: We propose a novel, lightweight time series prediction model, WaveTS-B.<n>This model combines wavelet transforms with a mechanism to capture both periodic and non-stationary characteristics of data in the wavelet domain.<n>We show that our model achieves state-of-the-art (SOTA) performance with significantly fewer parameters.
• Score: 7.478995447422547
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: The field of time series forecasting is rapidly advancing, with recent large-scale Transformers and lightweight Multilayer Perceptron (MLP) models showing strong predictive performance. However, conventional Transformer models are often hindered by their large number of parameters and their limited ability to capture non-stationary features in data through smoothing. Similarly, MLP models struggle to manage multi-channel dependencies effectively. To address these limitations, we propose a novel, lightweight time series prediction model, WaveTS-B. This model combines wavelet transforms with MLP to capture both periodic and non-stationary characteristics of data in the wavelet domain. Building on this foundation, we propose a channel clustering strategy that incorporates a Mixture of Experts (MoE) framework, utilizing a gating mechanism and expert network to handle multi-channel dependencies efficiently. We propose WaveTS-M, an advanced model tailored for multi-channel time series prediction. Empirical evaluation across eight real-world time series datasets demonstrates that our WaveTS series models achieve state-of-the-art (SOTA) performance with significantly fewer parameters. Notably, WaveTS-M shows substantial improvements on multi-channel datasets, highlighting its effectiveness.

Related papers

• WaveMoE: A Wavelet-Enhanced Mixture-of-Experts Foundation Model for Time Series Forecasting [43.52635148490259]
  We propose a new perspective that integrates explicit frequency-domain representations into scalable foundation models.<n>We introduce WaveMoE, a wavelet-enhanced mixture-of-experts foundation model for time series forecasting.<n>Preliminary experimental results on 16 diverse benchmark datasets indicate that WaveMoE has the potential to further improve forecasting performance.
  arXiv  Detail & Related papers  (2026-04-12T09:17:19Z)
• DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters [50.43534351968113]
  Existing generative time series models do not address the multi-dimensional properties of time series data well.<n>Inspired by Multimodal Diffusion Transformers that integrate textual guidance into video generation, we propose Diffusion Transformers for Time Series (DiTS)
  arXiv  Detail & Related papers  (2026-02-06T10:48:13Z)
• A Unified Frequency Domain Decomposition Framework for Interpretable and Robust Time Series Forecasting [81.73338008264115]
  Current approaches for time series forecasting, whether in the time or frequency domain, predominantly use deep learning models based on linear layers or transformers.<n>We propose FIRE, a unified frequency domain decomposition framework that provides a mathematical abstraction for diverse types of time series.<n>Fire consistently outperforms state-of-the-art models on long-term forecasting benchmarks.
  arXiv  Detail & Related papers  (2025-10-11T09:59:25Z)
• Multi-Scale Finetuning for Encoder-based Time Series Foundation Models [67.95907033226585]
  Time series foundation models (TSFMs) demonstrate impressive zero-shot performance for time series forecasting.<n>While naive finetuning can yield performance gains, we argue that it falls short of fully leveraging TSFMs' capabilities.<n>We propose Multiscale finetuning (MSFT), a simple yet general framework that explicitly integrates multi-scale modeling into the finetuning process.
  arXiv  Detail & Related papers  (2025-06-17T01:06:01Z)
• MFRS: A Multi-Frequency Reference Series Approach to Scalable and Accurate Time-Series Forecasting [51.94256702463408]
  Time series predictability is derived from periodic characteristics at different frequencies.<n>We propose a novel time series forecasting method based on multi-frequency reference series correlation analysis.<n> Experiments on major open and synthetic datasets show state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-03-11T11:40:14Z)
• WPMixer: Efficient Multi-Resolution Mixing for Long-Term Time Series Forecasting [20.29110166475336]
  We propose Wavelet Patch Patch Mixer (WPMixer), a novel multi-resolution wavelet decomposition model for time series forecasting.<n>Our model significantly outperforms state-of-the-art-based and transformer-based models for long-term time series forecasting.
  arXiv  Detail & Related papers  (2024-12-22T22:08:16Z)
• UniTST: Effectively Modeling Inter-Series and Intra-Series Dependencies for Multivariate Time Series Forecasting [98.12558945781693]
  We propose a transformer-based model UniTST containing a unified attention mechanism on the flattened patch tokens. Although our proposed model employs a simple architecture, it offers compelling performance as shown in our experiments on several datasets for time series forecasting.
  arXiv  Detail & Related papers  (2024-06-07T14:39:28Z)
• Adaptive Multi-Scale Decomposition Framework for Time Series Forecasting [26.141054975797868]
  We propose a novel Adaptive Multi-Scale Decomposition (AMD) framework for time series forecasting.<n>Our framework decomposes time series into distinct temporal patterns at multiple scales, leveraging the Multi-Scale Decomposable Mixing (MDM) block.<n>Our approach effectively models both temporal and channel dependencies and utilizes autocorrelation to refine multi-scale data integration.
  arXiv  Detail & Related papers  (2024-06-06T05:27:33Z)
• Unlocking the Power of Patch: Patch-Based MLP for Long-Term Time Series Forecasting [0.0]
  Recent studies have attempted to refine the Transformer architecture to demonstrate its effectiveness in Long-Term Time Series Forecasting tasks.<n>We attribute the effectiveness of these models largely to the adopted Patch mechanism.<n>We propose a novel and simple Patch-based components (PatchMLP) for LTSF tasks.
  arXiv  Detail & Related papers  (2024-05-22T12:12:20Z)
• Unified Training of Universal Time Series Forecasting Transformers [104.56318980466742]
  We present a Masked-based Universal Time Series Forecasting Transformer (Moirai) Moirai is trained on our newly introduced Large-scale Open Time Series Archive (LOTSA) featuring over 27B observations across nine domains. Moirai achieves competitive or superior performance as a zero-shot forecaster when compared to full-shot models.
  arXiv  Detail & Related papers  (2024-02-04T20:00:45Z)
• Timer: Generative Pre-trained Transformers Are Large Time Series Models [83.03091523806668]
  This paper aims at the early development of large time series models (LTSM) During pre-training, we curate large-scale datasets with up to 1 billion time points. To meet diverse application needs, we convert forecasting, imputation, and anomaly detection of time series into a unified generative task.
  arXiv  Detail & Related papers  (2024-02-04T06:55:55Z)
• Parsimony or Capability? Decomposition Delivers Both in Long-term Time Series Forecasting [46.63798583414426]
  Long-term time series forecasting (LTSF) represents a critical frontier in time series analysis. Our study demonstrates, through both analytical and empirical evidence, that decomposition is key to containing excessive model inflation. Remarkably, by tailoring decomposition to the intrinsic dynamics of time series data, our proposed model outperforms existing benchmarks.
  arXiv  Detail & Related papers  (2024-01-22T13:15:40Z)
• Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series [11.635608108358575]
  We introduce Tiny Time Mixers (TTM), a compact model with effective transfer learning capabilities, trained exclusively on public TS datasets. TTM incorporates innovations like adaptive patching, diverse resolution sampling, and resolution prefix tuning to handle pre-training on varied dataset resolutions. It outperforms existing popular benchmarks in zero/few-shot forecasting by (4-40%), while reducing computational requirements significantly.
  arXiv  Detail & Related papers  (2024-01-08T15:21:21Z)
• Multi-scale Transformer Pyramid Networks for Multivariate Time Series Forecasting [8.739572744117634]
  We introduce a dimension invariant embedding technique that captures short-term temporal dependencies. We present a novel Multi-scale Transformer Pyramid Network (MTPNet) specifically designed to capture temporal dependencies at multiple unconstrained scales.
  arXiv  Detail & Related papers  (2023-08-23T06:40:05Z)
• MTS-Mixers: Multivariate Time Series Forecasting via Factorized Temporal and Channel Mixing [18.058617044421293]
  This paper investigates the contributions and deficiencies of attention mechanisms on the performance of time series forecasting. We propose MTS-Mixers, which use two factorized modules to capture temporal and channel dependencies. Experimental results on several real-world datasets show that MTS-Mixers outperform existing Transformer-based models with higher efficiency.
  arXiv  Detail & Related papers  (2023-02-09T08:52:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.