Benutzer: Gast  Login
Titel:

Accelerating Molecular Graph Neural Networks via Knowledge Distillation

Dokumenttyp:
Konferenzbeitrag
Art des Konferenzbeitrags:
Textbeitrag / Aufsatz
Autor(en):
Ekström Kelvinius, Filip; Georgiev, Dimitar; Toshev, Artur; Gasteiger, Johannes
Abstract:
Recent advances in graph neural networks (GNNs) have enabled more comprehensive modeling of molecules and molecular systems, thereby enhancing the precision of molecular property prediction and molecular simulations. Nonetheless, as the field has been progressing to bigger and more complex architectures, state-of-the-art GNNs have become largely prohibitive for many large-scale applications. In this paper, we explore the utility of knowledge distillation (KD) for accelerating molecular GNNs. To...     »
Dewey-Dezimalklassifikation:
620 Ingenieurwissenschaften
Herausgeber:
Oh, A.; Neumann, T.; Globerson, A.; Saenko, K.; Hardt, M.; Levine, S.
Kongress- / Buchtitel:
Advances in Neural Information Processing Systems
Band / Teilband / Volume:
36
Verlag / Institution:
Curran Associates, Inc.
Jahr:
2023
Seiten:
25761--25792
Sprache:
en
WWW:
https://proceedings.neurips.cc/paper_files/paper/2023/file/51ec452ca04d8ec7160e5bbaf76153f6-Paper-Conference.pdf
TUM Einrichtung:
Lehrstuhl für Aerodynamik und Strömungsmechanik
 BibTeX