paper-with-me

Papers

SSDNet: State Space Decomposition Neural Network for Time Series Forecasting

2021-12-19 · Yang Lin, Irena Koprinska, Mashud Rana

In this paper, we present SSDNet, a novel deep learning approach for time series forecasting. SSDNet combines the Transformer architecture with state space models to provide probabilistic and interpretable forecasts, including trend and seasonality components and previous time steps important for the prediction. The Transformer architecture is used to learn the temporal patterns and estimate the parameters of the state space model directly and efficiently, without the need for Kalman filters. We comprehensively evaluate the performance of SSDNet on five data sets, showing that SSDNet is an effective method in terms of accuracy and speed, outperforming state-of-the-art deep learning and statistical methods, and able to provide meaningful trend and seasonality components.

📄 PDF Abstract BibTeX arXiv:2112.10251

Code (1)

YangLIN1997/SSDNet-ICDM2021 공식 구현 pytorch

Tasks

Deep LearningState Space ModelsTime SeriesTime Series AnalysisTime Series Forecasting

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Position-Wise Feed-Forward Layer 설명 없음
Adam 설명 없음
Residual Connection 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Similar Papers 제목 키워드 기반

Spherical Space Feature Decomposition for Guided Depth Map Super-Resolution

2023-03-15 · ICCV 2023 1 · Zixiang Zhao, Jiangshe Zhang, Xiang Gu, Chengli Tan 외

Guided depth map super-resolution (GDSR), as a hot topic in multi-modal image processing, aims to upsample low-resolution (LR) depth maps with additional information involved in high-resolution (HR) RGB images from the s…

Contrastive LearningDepth Map Super-ResolutionSuper-Resolution

A Single Image Is All You Need: Zero-Shot Anomaly Localization Without Training Data

2025-09-22 · Mehrdad Moradi, Shengzhe Chen, Hao Yan, Kamran Paynabar arxiv

Anomaly detection in images is typically addressed by learning from collections of training data or relying on reference samples. In many real-world scenarios, however, such training data may be unavailable, and only the…

Anomaly Detection

Kolmogorov Space in Time Series Data

2016-06-10

We provide the proof that the space of time series data is a Kolmogorov space with $T_{0}$-separation axiom using the loop space of time series data. In our approach we define a cyclic coordinate of intrinsic time scale …

Time SeriesTime Series Analysis

Memory-Efficient Centroid Decomposition for Long Time Series

2017-04-01 · IEEE 30th International Conference on Data Engineering (ICDE) 2017 4 · Mourad Khayati, Michael Böhlen, Johann Gamper

Real-world applications that deal with time series data often rely on matrix decomposition techniques, such as the Singular Value Decomposition (SVD). The Centroid Decomposition (CD) approximates the Singular Value Decom…

Time SeriesTime Series Analysis

Linear Simple Cycle Reservoirs at the edge of stability perform Fourier decomposition of the input driving signals

2024-11-30 · Robert Simon Fong, Boyu Li, Peter Tino

This paper explores the representational structure of linear Simple Cycle Reservoirs (SCR) operating at the edge of stability. We view SCR as providing in their state space feature representations of the input-driving ti…

Time Series