tml-epfl/why-weight-decay
Why Do We Need Weight Decay in Modern Deep Learning? [NeurIPS 2024]
PythonNOASSERTION
Stargazers
- 152334HNational University of Singapore
- alec-hoyland@YurtsAI
- cweiqiangAI Singapore
- cyrtaMetamedia Technologies
- deburgur
- denisfitz57
- Dingding-Han
- dngfra
- EmreOzkoseHacettepe University
- fly51flyPRIS
- GONGXI1994
- hardikudeshi
- Hiroki11xMila, Université de Montréal
- jamal-ansaryUniversity of Toledo
- JeffCarpenterCanada
- MarcellusZhaoÉcole Polytechnique Fédérale de Lausanne
- max-andrEPFL
- MiralanShanghai, China
- misumisumiJapan
- MonoHueShanghai, China
- progerSupercomputer City
- qmdnlsSeoul, Korea
- radarFudanNUS
- robflynnyhThe University of Sheffield
- Ryu1845
- skandermoalla@CLAIRE-Labo EPFL
- speedcell4NICT
- sustcsonglinMIT
- SynapticSageBrandeis University
- ternausternaus.blog
- tlin-taolin@epfml
- VincentqywTHU
- whlzyShanghai Jiao Tong University
- wuriningUK
- yaodongyuUC Berkeley
- yzhangcsSoochow University