Prati
John Chen
Naslov
Citirano
Citirano
Godina
Demon: improved neural network training with momentum decay
J Chen, C Wolfe, Z Li, A Kyrillidis
ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and …, 2022
31*2022
Negative sampling in semi-supervised learning
J Chen, V Shah, A Kyrillidis
ICML 2020, 2020
232020
StackMix: A complementary Mix algorithm
J Chen, S Sinha, A Kyrillidis
UAI 2022, 2022
15*2022
REX: Revisiting Budgeted Training with an Improved Schedule
J Chen, C Wolfe, A Kyrillidis
MLSys 2022, 2021
92021
Mitigating deep double descent by concatenating inputs
J Chen, Q Wang, A Kyrillidis
ICML 2021 Overparameterization: Pitfalls & Opportunities Workshop. CIKM 2021., 2021
42021
STORM: foundations of end-to-end empirical risk minimization on the edge
B Coleman, G Gupta, J Chen, A Shrivastava
arXiv preprint arXiv:2006.14554, 2020
42020
Revisiting Consistent Hashing with Bounded Loads
J Chen, B Coleman, A Shrivastava
AAAI 2021, 2021
22021
Fast FixMatch: Faster Semi-Supervised Learning with Curriculum Batch Size
J Chen, C Dun, A Kyrillidis
arXiv preprint arXiv:2309.03469, 2023
2023
Sustav trenutno ne može provesti ovu radnju. Pokušajte ponovo kasnije.
Članci 1–8