Timezone: »

GradInit: Learning to Initialize Neural Networks for Stable and Efficient Training
Chen Zhu · Renkun Ni · Zheng Xu · Kezhi Kong · W. Ronny Huang · Tom Goldstein

Wed Dec 08 12:30 AM -- 02:00 AM (PST) @

Innovations in neural architectures have fostered significant breakthroughs in language modeling and computer vision. Unfortunately, novel architectures often result in challenging hyper-parameter choices and training instability if the network parameters are not properly initialized. A number of architecture-specific initialization schemes have been proposed, but these schemes are not always portable to new architectures. This paper presents GradInit, an automated and architecture agnostic method for initializing neural networks. GradInit is based on a simple heuristic; the norm of each network layer is adjusted so that a single step of SGD or Adam with prescribed hyperparameters results in the smallest possible loss value. This adjustment is done by introducing a scalar multiplier variable in front of each parameter block, and then optimizing these variables using a simple numerical scheme. GradInit accelerates the convergence and test performance of many convolutional architectures, both with or without skip connections, and even without normalization layers. It also improves the stability of the original Transformer architecture for machine translation, enabling training it without learning rate warmup using either Adam or SGD under a wide range of learning rates and momentum coefficients. Code is available at https://github.com/zhuchen03/gradinit.

Author Information

Chen Zhu (University of Maryland, College Park)
Renkun Ni (University of Maryland)
Zheng Xu (Google AI)
Kezhi Kong (University of Maryland, College Park)
W. Ronny Huang (Google Research)
Tom Goldstein (Rice University)

More from the Same Authors