The Unreasonable Effectiveness Of Early Discarding After One Epoch In Neural Network Hyperparameter Optimization

arXiv:2404.04111v1 Announce Type: new
Abstract: To reach high performance with deep learning, hyperparameter optimization (HPO) is essential. This process is usually time-consuming due to costly evaluations of neural networks. Early discarding techniques limit the resources granted to unpromising candidates by observing the empirical learning curves and canceling neural network training as soon as the lack of competitiveness of a candidate becomes evident. Despite two decades of research, little is understood about the trade-off between the aggressiveness of discarding and the …

abstract arxiv cs.lg deep learning hyperparameter network networks neural network neural networks optimization performance process resources type