Researcher on Robust Graph Machine Learning
PhD Student @ Technical University of Munich, DAML Group & relAI Fellow.
Quirk for Philosphy. He/Him
| Website | https://saper0.github.io/ |
You can sample nodes for scalable #GNN #training. But how do you do #scalable #inference?
In our latest paper (Oral #LogConference
) we introduce influence-based mini-batching (#IBMB) for both fast inference and training, achieving up to 130x and 17x speedups, respectively!
1/8 in 🧵
Hi all, I research the #robustness of #graph learning models doing a mix of #theory and #practice.
Currently, I am a PhD Student @ Technical University of Munich #TUM and a #relAI Fellow. You can find more information on my website: https://saper0.github.io/.
If you are interested in my work, always feel free to reach out :).
Johannes Gasteiger