arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2008.12478cs.LGstat.ML
收藏

预测训练时间而不训练

Predicting Training Time Without Training

University of Padova(帕多瓦大学);Amazon Web Services(亚马逊云科技)

Luca Zancato, Alessandro Achille, Avinash Ravichandran, Rahul Bhotika, Stefano Soatto

摘要:

我们解决了预测预训练的深度网络需要收敛到损失函数的给定值的优化步骤的数量的问题。为了做到这一点,我们利用了这样一个事实,即在微调过程中,深度网络的训练动态可以很好地近似于线性化模型的训练动态。这使我们能够通过在函数空间中求解低维随机微分方程(Stochastic Differential Equation,简称SDD)来近似训练过程中任何一点的训练损失和准确度。使用这个结果,我们能够预测随机梯度下降(SGD)将模型微调到给定损失所需的时间,而无需执行任何训练。在我们的实验中,我们能够在各种数据集和超参数上预测ResNet的训练时间,误差幅度在20%以内,与实际训练相比,成本降低了30到45倍。我们还讨论了如何进一步降低我们方法的计算和内存成本,特别是我们表明,通过利用梯度矩阵的谱特性,可以在只处理样本子集的情况下预测大型数据集的训练时间。

英文摘要:

We tackle the problem of predicting the number of optimization steps that a pre-trained deep network needs to converge to a given value of the loss function. To do so, we leverage the fact that the training dynamics of a deep network during fine-tuning are well approximated by those of a linearized model. This allows us to approximate the training loss and accuracy at any point during training by solving a low-dimensional Stochastic Differential Equation (SDE) in function space. Using this result, we are able to predict the time it takes for Stochastic Gradient Descent (SGD) to fine-tune a model to a given loss without having to perform any training. In our experiments, we are able to predict training time of a ResNet within a 20% error margin on a variety of datasets and hyper-parameters, at a 30 to 45-fold reduction in cost compared to actual training. We also discuss how to further reduce the computational and memory cost of our method, and in particular we show that by exploiting the spectral properties of the gradients' matrix it is possible predict training time on a large dataset while processing only a subset of the samples.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑