<ul><li>Dynamic graph learning requires effective modeling of temporal relationships in applications involving temporal networks.</li><li>This paper explores the use of linear time encoder as an alternative to sinusoidal time encoder.</li><li>The linear time encoder avoids temporal information loss caused by sinusoidal functions and reduces the need for high dimensional time encoders.</li><li>Experimental results show that the linear time encoder improves the performance of existing models and leads to significant parameter savings.</li></ul>

Between Linear and Sinusoidal: Rethinking the Time Encoder in Dynamic Graph Learning

Discover more