Luigi Acerbi, Samuel Kaski, Markus Heinonen, Trung Trinh
We lifted 9 functions out of this paper's own repositories and ran 7 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| trungtrinh44/damp | canonical | 7 of 9 |
| Function | Status | Where it lives |
|---|---|---|
| _get_gaussian_kernel | Ran | trungtrinh44/damp/randaugment/augment.py pointer only (licence: NONE) · get_code("ed852e74db36309a") |
| _get_gaussian_kernel_2d | Ran | trungtrinh44/damp/randaugment/augment.py pointer only (licence: NONE) · get_code("03fcb094182e80ea") |
| _normalize_tuple | Ran | trungtrinh44/damp/randaugment/augment.py pointer only (licence: NONE) · get_code("a5c06a4fe76d4fc3") |
| _pad | Ran | trungtrinh44/damp/randaugment/augment.py pointer only (licence: NONE) · get_code("f55c98808a12d76e") |
| gaussian_filter2d | Ran | trungtrinh44/damp/randaugment/augment.py pointer only (licence: NONE) · get_code("c15f0fd201ea7a3a") |
| open_json | Ran | trungtrinh44/damp/imagenet/eval_vit_imagenet_and_imagenet_c.py pointer only (licence: NONE) · get_code("a1c31b015211f700") |
| to_4d | Ran | trungtrinh44/damp/randaugment/augment.py pointer only (licence: NONE) · get_code("bd1c72a9d8556505") |
| from_4d | Not yet run | trungtrinh44/damp/randaugment/augment.py pointer only (licence: NONE) · get_code("b10fb8ac53c53a97") |
| gaussian_noise | Not yet run | trungtrinh44/damp/randaugment/augment.py pointer only (licence: NONE) · get_code("8cbf8b03fdf303a9") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Deep neural networks (DNNs) excel on clean images but struggle with corrupted ones. Incorporating specific corruptions into the data augmentation pipeline can improve robustness to those corruptions but may harm performance on clean images and other types of distortion. In this paper, we introduce an alternative approach that improves the robustness of DNNs to a wide range of corruptions without compromising accuracy on clean images. We first demonstrate that input perturbations can be mimicked by multiplicative perturbations in the weight space. Leveraging this, we propose Data Augmentation via Multiplicative Perturbation (DAMP), a training method that optimizes DNNs under random multiplicative weight perturbations. We also examine the recently proposed Adaptive Sharpness-Aware Minimization (ASAM) and show that it optimizes DNNs under adversarial multiplicative weight perturbations. Experiments on image classification datasets (CIFAR-10/100, Tiny-ImageNet and ImageNet) and neural network architectures (ResNet50, ViT-S/16, ViT-B/16) show that DAMP enhances model generalization performance in the presence of corruptions across different settings. Notably, DAMP is able to train a ViT-S/16 on ImageNet from scratch, reaching the top-1 error of 23.7% which is comparable to ResNet50 without extensive data augmentations. 1
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2406.16540")
get_code_for_paper("2406.16540")
have("2406.16540")
Connect an agent — have() is free.