Направление градиента
otherentity
ID: gradientdirection
Related Papers
Dynamically Adjusting Transformer Batch Size by Monitoring Gradient Direction Change
AbstractThe choice of hyper-parameters affects the performance of neural models. While much previous research (Sutskever et al., 2013; Duchi et al., 2011; Kingma and Ba, 2015) focuses on accelerating convergence …...