CS Compiler Seminar: Please join us on 10/28/2024, from 4:30-5:30pm in the Room 2406 (Siebel) where Wanyu Zhao will give a talk, “TGLite: A Lightweight Programming Framework for Continuous-Time Temporal Graph Neural Networks”. Please see their abstract below:
Speaker(s): Wanyu Zhao
Title: TGLite: A Lightweight Programming Framework for Continuous-Time Temporal Graph Neural Networks
Conference: ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '24)
Author(s): Yufeng Wang (UIUC), Charith Mendis (UIUC)
Abstract: In recent years, Temporal Graph Neural Networks (TGNNs) have achieved great success in learning tasks for graphs that change over time. These dynamic/temporal graphs repre- sent topology changes as either discrete static graph snap- shots (called DTDGs), or a continuous stream of timestamped edges (called CTDGs). Because continuous-time graphs have richer time information, it will be crucial to have abstractions for programming CTDG-based models so that practitioners can easily explore new designs and optimizations in this space. A few recent frameworks have been proposed for pro- gramming and accelerating TGNN models, but these either do not support continuous-time graphs, lack easy compos- ability, and/or do not facilitate CTDG-specific optimizations.