Resolving learning rates adaptively by locating stochastic non-negative associated gradient projection points using line searches
Sign Up to like & getrecommendations! Published in 2021 at "Journal of Global Optimization"
DOI: 10.1007/s10898-020-00921-z
Abstract: Learning rates in stochastic neural network training are currently determined a priori to training, using expensive manual or automated iterative tuning. Attempts to resolve learning rates adaptively, using line searches, have proven computationally demanding. Reducing… read more here.
Keywords: rates adaptively; line; learning rates; line searches ... See more keywords