2 papers
cs.LG2026
AXLearn: Modular, Hardware-Agnostic Large Model Training
Mark Lee, Chang Lan, Tom Gunter +34
AXLearn is a production system which facilitates scalable and high-performance training of large deep learning models. Compared to other state-of-art deep learning systems, AXLearn…
math.AG2024
Toric Fano manifolds that do not admit extremal Kähler metrics
DongSeon Hwang, Hiroshi Sato, Naoto Yotsutani
We show that there exists a toric Fano manifold of dimension that does not admit an extremal Kähler metric in the first Chern class, answering a question of Mabuchi. By takin…