Diagnostic Spatio-temporal Transformer with Faithful Encoding

by   Jokin Labaien, et al.

This paper addresses the task of anomaly diagnosis when the underlying data generation process has a complex spatio-temporal (ST) dependency. The key technical challenge is to extract actionable insights from the dependency tensor characterizing high-order interactions among temporal and spatial indices. We formalize the problem as supervised dependency discovery, where the ST dependency is learned as a side product of multivariate time-series classification. We show that temporal positional encoding used in existing ST transformer works has a serious limitation in capturing higher frequencies (short time scales). We propose a new positional encoding with a theoretical guarantee, based on discrete Fourier transform. We also propose a new ST dependency discovery framework, which can provide readily consumable diagnostic information in both spatial and temporal directions. Finally, we demonstrate the utility of the proposed model, DFStrans (Diagnostic Fourier-based Spatio-temporal Transformer), in a real industrial application of building elevator control.


page 17

page 18

page 27


Clustering of Time Series Data with Prior Geographical Information

Time Series data are broadly studied in various domains of transportatio...

Stecformer: Spatio-temporal Encoding Cascaded Transformer for Multivariate Long-term Time Series Forecasting

Multivariate long-term time series forecasting is of great application a...

Spatio-Temporal Transformer-Based Reinforcement Learning for Robot Crowd Navigation

The social robot navigation is an open and challenging problem. In exist...

Epileptic Seizure Classification with Symmetric and Hybrid Bilinear Models

Epilepsy affects nearly 1 be treated by anti-epileptic drugs and a much ...

A new class of spatial covariance functions generated by higher-order kernels

Covariance functions and variograms play a fundamental role in explorato...

Trajectory annotation using sequences of spatial perception

In the near future, more and more machines will perform tasks in the vic...

Crime Linkage Detection by Spatial-Temporal-Textual Point Processes

Crimes emerge out of complex interactions of behaviors and situations; t...

Please sign up or login with your details

Forgot password? Click here to reset