.png){kind=icon}
Return
Asynchronous graph generator
.png){kind=icon}
OA
AI
DOI：10.1016/j.sigpro.2025.110183.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A novel interpretation of sparse multi-channel time series as an asynchronous graph, where nodes represent the observations. • Novel use of embeddings to encode temporal and channel features which can be leveraged through graph attention to model the relationships among observations. • The introduction of conditional attention generation, a mechanism to generate new observations conditioned on given temporal/channel features. • An experimental validation of the AGG (Asynchronous Graph Generator) on standard benchmarks against the state of the art. • A study of the limiting performance of the AGG from the perspective of data augmentation.
Journal
IF:
3.6
Papers:
10.0K
Citations:
1.7W
Organization
No organization information available
Cited Papers
ShareSaveShareSaveTraffic transformer: Capturing the continuity and periodicity of time series for traffic forecasting
PREAI
ShareSaveConnecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks.png)
Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
OAAI
Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data MiningIF0
OAAIShareSaveShareSave
View more