Sensorformer: Cross-patch attention with global-patch compression is effective for high-dimensional multivariate time series forecasting

• URL: http://arxiv.org/abs/2501.03284v1
• Date: Mon, 06 Jan 2025 03:14:47 GMT
• Title: Sensorformer: Cross-patch attention with global-patch compression is effective for high-dimensional multivariate time series forecasting
• Authors: Liyang Qin, Xiaoli Wang, Chunhua Yang, Huaiwen Zou, Haochuan Zhang,
• Abstract summary: We propose a new Transformer, Sensorformer, which first compresses the global patch information and then simultaneously extracts cross-variable and cross-time dependencies from the compressed representations.<n>Sensorformer can effectively capture the correct inter-variable correlations and causal relationships, even in the presence of dynamic causal lags between variables.
• Score: 12.103678233732584
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Among the existing Transformer-based multivariate time series forecasting methods, iTransformer, which treats each variable sequence as a token and only explicitly extracts cross-variable dependencies, and PatchTST, which adopts a channel-independent strategy and only explicitly extracts cross-time dependencies, both significantly outperform most Channel-Dependent Transformer that simultaneously extract cross-time and cross-variable dependencies. This indicates that existing Transformer-based multivariate time series forecasting methods still struggle to effectively fuse these two types of information. We attribute this issue to the dynamic time lags in the causal relationships between different variables. Therefore, we propose a new multivariate time series forecasting Transformer, Sensorformer, which first compresses the global patch information and then simultaneously extracts cross-variable and cross-time dependencies from the compressed representations. Sensorformer can effectively capture the correct inter-variable correlations and causal relationships, even in the presence of dynamic causal lags between variables, while also reducing the computational complexity of pure cross-patch self-attention from $O(D^2 \cdot Patch\_num^2 \cdot d\_model)$ to $O(D^2 \cdot Patch\_num \cdot d\_model)$. Extensive comparative and ablation experiments on 9 mainstream real-world multivariate time series forecasting datasets demonstrate the superiority of Sensorformer. The implementation of Sensorformer, following the style of the Time-series-library and scripts for reproducing the main results, is publicly available at https://github.com/BigYellowTiger/Sensorformer

Related papers

• Gateformer: Advancing Multivariate Time Series Forecasting through Temporal and Variate-Wise Attention with Gated Representations [2.2091590689610823]
  We re-purpose the Transformer architecture to model both cross-time and cross-variate dependencies. Our method achieves state-of-the-art performance across 13 real-world datasets, delivering performance improvements up to 20.7% over original models.
  arXiv  Detail & Related papers  (2025-05-01T04:59:05Z)
• Decomposition-based multi-scale transformer framework for time series anomaly detection [0.9438207505148947]
  We propose a transformer-based framework built on decomposition (TransDe) for time series anomaly detection. A multi-scale patch-based transformer architecture is proposed to exploit the representative dependencies of each decomposed component of the time series. A novel asynchronous loss function with a stop-gradient strategy is introduced to enhance the performance of TransDe effectively.
  arXiv  Detail & Related papers  (2025-04-19T06:47:38Z)
• Sentinel: Multi-Patch Transformer with Temporal and Channel Attention for Time Series Forecasting [48.52101281458809]
  Transformer-based time series forecasting has recently gained strong interest due to the ability of transformers to model sequential data. We propose Sentinel, a transformer-based architecture composed of an encoder able to extract contextual information from the channel dimension. We introduce a multi-patch attention mechanism, which leverages the patching process to structure the input sequence in a way that can be naturally integrated into the transformer architecture.
  arXiv  Detail & Related papers  (2025-03-22T06:01:50Z)
• Timer-XL: Long-Context Transformers for Unified Time Series Forecasting [67.83502953961505]
  We present Timer-XL, a generative Transformer for unified time series forecasting.<n>Timer-XL achieves state-of-the-art performance across challenging forecasting benchmarks through a unified approach.
  arXiv  Detail & Related papers  (2024-10-07T07:27:39Z)
• 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)
• Leveraging 2D Information for Long-term Time Series Forecasting with Vanilla Transformers [55.475142494272724]
  Time series prediction is crucial for understanding and forecasting complex dynamics in various domains. We introduce GridTST, a model that combines the benefits of two approaches using innovative multi-directional attentions. The model consistently delivers state-of-the-art performance across various real-world datasets.
  arXiv  Detail & Related papers  (2024-05-22T16:41:21Z)
• VCformer: Variable Correlation Transformer with Inherent Lagged Correlation for Multivariate Time Series Forecasting [1.5165632546654102]
  We propose Variable Correlation Transformer (VCformer) to mine the correlations among variables. VCA calculates and integrates the cross-correlation scores corresponding to different lags between queries and keys. Inspired by Koopman dynamics theory, we also develop Koopman Temporal Detector (KTD) to better address the non-stationarity in time series.
  arXiv  Detail & Related papers  (2024-05-19T07:39:22Z)
• EdgeConvFormer: Dynamic Graph CNN and Transformer based Anomaly Detection in Multivariate Time Series [7.514010315664322]
  We propose a novel anomaly detection method, named EdgeConvFormer, which integrates stacked Time2vec embedding, dynamic graph CNN, and Transformer to extract global and local spatial-time information. Experiments demonstrate that EdgeConvFormer can learn the spatial-temporal modeling from multivariate time series data and achieve better anomaly detection performance than the state-of-the-art approaches on many real-world datasets of different scales.
  arXiv  Detail & Related papers  (2023-12-04T08:38:54Z)
• iTransformer: Inverted Transformers Are Effective for Time Series Forecasting [62.40166958002558]
  We propose iTransformer, which simply applies the attention and feed-forward network on the inverted dimensions. The iTransformer model achieves state-of-the-art on challenging real-world datasets.
  arXiv  Detail & Related papers  (2023-10-10T13:44:09Z)
• CARD: Channel Aligned Robust Blend Transformer for Time Series Forecasting [50.23240107430597]
  We design a special Transformer, i.e., Channel Aligned Robust Blend Transformer (CARD for short), that addresses key shortcomings of CI type Transformer in time series forecasting. First, CARD introduces a channel-aligned attention structure that allows it to capture both temporal correlations among signals. Second, in order to efficiently utilize the multi-scale knowledge, we design a token blend module to generate tokens with different resolutions. Third, we introduce a robust loss function for time series forecasting to alleviate the potential overfitting issue.
  arXiv  Detail & Related papers  (2023-05-20T05:16:31Z)

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.