# RobustDBNN

**Repository Path**: wdz2020/robust-dbnn

## Basic Information

- **Project Name**: RobustDBNN
- **Description**: This is a code of ICML-25 for the study of 'Can DBNNs Robust to Environmental Noise for Resource-constrained Scenarios?'.
- **Primary Language**: Python
- **License**: CC-BY-4.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-05-11
- **Last Updated**: 2025-05-11

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# RobustDBNN

#### Introduction
This is a code of ICML-25 for the study of 'Can DBNNs Robust to Environmental Noise for Resource-constrained Scenarios?'.

#### Equipment environment
For the Ubuntu 18.04, the GPU can be 3090, 1080ti, or A800.

#### Main Environment Version
PyTorch=1.12
Python=3.9.X

#### Dataset
[The download URL address of the Brain Tumor dataset](https://www.kaggle.com/datasets/masoudnickparvar/brain-tumor-mri-dataset)


#### Usage
For Example: 

You can run the Brain_Tumor_classification_cyclebnn_our_resnet34_noise_image_norm_0.05_0.1.py file directly to implement the Brain Tumor classification task.
Then, you can execute the following command:
```
python Brain_Tumor_classification_cyclebnn_our_resnet34_noise_image_norm_0.05_0.1.py
```

#### References
If you are interested, please cite this paper (Thank you very much!).

@inproceedings{RobustDBNN,
	author={Wendong Zheng, Junyang Chen, Husheng Guo and Wenjian Wang},
	title={Can DBNNs Robust to Environmental Noise for Resource-constrained Scenarios?},
	booktitle    = {The Proceedings of the International Conference on Machine Learning (ICML)},
	year={2025} 
}

#### Others
More experimental codes will be systematically organized and uploaded in the future.