[R] AutoInit Software for Model Initialization

AutoInit is a weight initialization method that automatically adapts to different neural network architectures. It tracks the mean and variance of signals as they propagate through the network and initializes the weights at each layer to avoid exploding or vanishing signals. AutoInit can be used to improve performance of feedforward, convolutional, and residual networks; configured with different activation function, dropout, weight decay, learning rate, and normalizer settings; and applied to vision, language, tabular, multi-task, and transfer learning domains. The software …

machinelearning software