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.
Torch.manual_seed(3407) is all you need: On the influence of random seeds in deep learning architectures for computer vision
The investigation into random seed selection reveals significant variability in model accuracy, even with non-large variance, across the CIFAR 10 and Imagenet datasets.
- Year
- 2021
- Venue
- arXiv 2021
- Authors
- 1
- Hosting
- Abstract onlyARXIV-DEFAULT
Cite
Notes
Only stored in your browser.
Attribution
- Abstract & full text
- arxiv.org/abs/2109.08203v2ARXIV-DEFAULT
- TL;DR
- Semantic Scholar