arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2109.08203cs.CV
收藏

Torch.manual_seed(3407) 是你所需要的一切:论随机种子在计算机视觉深度学习架构中的影响

Torch.manual_seed(3407) is all you need: On the influence of random seeds in deep learning architectures for computer vision

École des Ponts(巴黎路桥学院)

David Picard

AI总结:

本文探究随机种子对计算机视觉深度学习模型准确率的影响,通过大规模种子扫描发现异常值易出现,性能波动显著。

摘要:

本文研究了随机种子选择对使用流行深度学习架构进行计算机视觉任务时准确率的影响。我在CIFAR-10上扫描了大量种子(多达$10^4$个),并在ImageNet上使用预训练模型扫描了较少种子,以探究大规模数据集的情况。结论是,即使方差不是很大,也令人惊讶地容易找到表现远优于或远差于平均值的异常值。

英文摘要:

In this paper I investigate the effect of random seed selection on the accuracy when using popular deep learning architectures for computer vision. I scan a large amount of seeds (up to $10^4$) on CIFAR 10 and I also scan fewer seeds on Imagenet using pre-trained models to investigate large scale datasets. The conclusions are that even if the variance is not very large, it is surprisingly easy to find an outlier that performs much better or much worse than the average.

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。

↑