Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

FTS: A Framework to Find a Faithful TimeSieve

Songning Lai, Ninghui Feng, Haochen Sui, Ze Ma, Hao Wang, Zichen Song, Hang Zhao, Yutao Yue

TL;DR

The paper tackles faithfulness in time-series forecasting by exposing TimeSieve's sensitivity to seeds and input perturbations and proposing Faithful TimeSieve (FTS). It formalizes FTS via three criteria—Similarity in IB Space, Closeness of Forecasting, and Stability of Forecasting—and introduces a minimax framework with auxiliary stability losses to enforce robustness. Empirical results on an Exchange Rate dataset show that FTS maintains or improves original-case accuracy while dramatically reducing sensitivity to seeds and input noise, advancing reliability in temporal forecasting. The work offers a pathway to generalize faithfulness to other temporal models, contributing to trustworthy decision-making in domains relying on time-series predictions.

Abstract

The field of time series forecasting has garnered significant attention in recent years, prompting the development of advanced models like TimeSieve, which demonstrates impressive performance. However, an analysis reveals certain unfaithfulness issues, including high sensitivity to random seeds and minute input noise perturbations. Recognizing these challenges, we embark on a quest to define the concept of \textbf{\underline{F}aithful \underline{T}ime\underline{S}ieve \underline{(FTS)}}, a model that consistently delivers reliable and robust predictions. To address these issues, we propose a novel framework aimed at identifying and rectifying unfaithfulness in TimeSieve. Our framework is designed to enhance the model's stability and resilience, ensuring that its outputs are less susceptible to the aforementioned factors. Experimentation validates the effectiveness of our proposed framework, demonstrating improved faithfulness in the model's behavior. Looking forward, we plan to expand our experimental scope to further validate and optimize our algorithm, ensuring comprehensive faithfulness across a wide range of scenarios. Ultimately, we aspire to make this framework can be applied to enhance the faithfulness of not just TimeSieve but also other state-of-the-art temporal methods, thereby contributing to the reliability and robustness of temporal modeling as a whole.

FTS: A Framework to Find a Faithful TimeSieve

TL;DR

The paper tackles faithfulness in time-series forecasting by exposing TimeSieve's sensitivity to seeds and input perturbations and proposing Faithful TimeSieve (FTS). It formalizes FTS via three criteria—Similarity in IB Space, Closeness of Forecasting, and Stability of Forecasting—and introduces a minimax framework with auxiliary stability losses to enforce robustness. Empirical results on an Exchange Rate dataset show that FTS maintains or improves original-case accuracy while dramatically reducing sensitivity to seeds and input noise, advancing reliability in temporal forecasting. The work offers a pathway to generalize faithfulness to other temporal models, contributing to trustworthy decision-making in domains relying on time-series predictions.

Abstract

The field of time series forecasting has garnered significant attention in recent years, prompting the development of advanced models like TimeSieve, which demonstrates impressive performance. However, an analysis reveals certain unfaithfulness issues, including high sensitivity to random seeds and minute input noise perturbations. Recognizing these challenges, we embark on a quest to define the concept of \textbf{\underline{F}aithful \underline{T}ime\underline{S}ieve \underline{(FTS)}}, a model that consistently delivers reliable and robust predictions. To address these issues, we propose a novel framework aimed at identifying and rectifying unfaithfulness in TimeSieve. Our framework is designed to enhance the model's stability and resilience, ensuring that its outputs are less susceptible to the aforementioned factors. Experimentation validates the effectiveness of our proposed framework, demonstrating improved faithfulness in the model's behavior. Looking forward, we plan to expand our experimental scope to further validate and optimize our algorithm, ensuring comprehensive faithfulness across a wide range of scenarios. Ultimately, we aspire to make this framework can be applied to enhance the faithfulness of not just TimeSieve but also other state-of-the-art temporal methods, thereby contributing to the reliability and robustness of temporal modeling as a whole.
Paper Structure (17 sections, 10 equations, 5 figures, 2 tables)

This paper contains 17 sections, 10 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Time Series Forecasting
  4. Faithful Time Series
  5. Method
  6. Preliminaries: TimeSieve
  7. Wavelet Decomposition and Reconstruction
  8. Information Filtering and Compression
  9. Faithfulness issues in TimeSieve
  10. What is a "faithful time series forecasting"?
  11. Defination for FTS
  12. Faithful TimeSieve Framework
  13. Eeperiments
  14. Datasets
  15. Settings
  16. ...and 2 more sections

Figures (5)

  • Figure 1: Visualization of the performance of the TimeSieve model versus other state-of-the-art models on the Exchange dataset. The left ordinate represents the MAE loss of the model, and the right ordinate represents the percentage of improved performance of the model over Autoformer.
  • Figure 2: Five random seeds [2021,2022,2023,2024,2025] are selected to train TimeSieve respectively. With "2021" as the base model, the percentage of variation is calculated and showed.
  • Figure 3: Framework of our proposed Faithful TimeSieve (FTS)
  • Figure 4: This figure presents a performance comparison of various advanced models, under conditions without any added perturbations. The left ordinate represents the MAE loss of the model, and the right ordinate represents the percentage of improved performance of the model over Autoformer.
  • Figure 5: Five random seeds [2021,2022,2023,2024,2025] are selected to train Faithful TimeSieve respectively. With "2021" as the base model, the percentage of variation is calculated and showed.

Theorems & Definitions (1)

  • Definition 1: Faithful TimeSieve (FTS)