Training feed-forward neural networks using the gradient descent method with the optimal stepsize [Elektronisk resurs] [PDF]
The most widely used algorithm for training multiplayer feedforward networks, Error BackPropagation (EBP), is an iterative gradient descend algorithm by nature. Variable stepsize is the key to fast convergence of BP networks.
Li, Yanming, +3 more
core
Fairness in the use of limited resources during a pandemic. [PDF]
Schosser J.
europepmc +1 more source
On the finding of stepsize in the numerical integration of the initial value problem [PDF]
Bu çalışmada, Cauchy probleminin nümerik integrasyonu için Picard teoremi tabanlı değişken adım genişliği seçimi ve hata analizi tabanlı değişken adım genişliği seçimi elde edilmiştir. Bu seçimlere bağlı olarak adım genişliği ve yaklaşık çözüm hesaplayan
Kızılkan, Gülnur Çelik
core
An efficient numerical method on modified space-time sparse grid for time-fractional diffusion equation with nonsmooth data. [PDF]
Zhu BY, Xiao AG, Li XY.
europepmc +1 more source
Large Stepsize Gradient Descent for Logistic Loss: Non-Monotonicity of the Loss Improves Optimization Efficiency [PDF]
We consider gradient descent (GD) with a constant stepsize applied to logistic regression with linearly separable data, where the constant stepsize $\eta$ is so large that the loss initially oscillates.
Wu, Jingfeng +3 more
core +1 more source
Music Sound Quality Assessment in Bimodal Cochlear Implant Users-Toward Improved Hearing Aid Fitting. [PDF]
Abdellatif KHA +5 more
europepmc +1 more source
Discrete stochastic maximal regularity. [PDF]
Evangelopoulos-Ntemiris F, Veraar M.
europepmc +1 more source
Quantization avoids saddle points in distributed optimization. [PDF]
Bo Y, Wang Y.
europepmc +1 more source
Efficient EM Estimation for the Pogit Model via Polya-Gamma Augmentation. [PDF]
Gutiérrez I, Ramírez S, Jofré L.
europepmc +1 more source
Seroresponse to repeated infections with Salmonella enterica Typhi and Paratyphi A. [PDF]
Teunis PFM +8 more
europepmc +1 more source

