We're trying to make model better which is robust against to adversarial images, especially made by FGSM. Yann LeCun's MNIST datasets are used.
We're inspired by this tutorial.
- train model with original MNIST datasets (learning rate == 0.001)
- get adversarial images of MNIST from trained model
- fine-tune model with adversarial images. learning rate is 0.0001 (it may be modified)
- validate with validation set 100 epochs each models
- results saved as a plot
A function named
generate_image_adversarial(args) is just interpretation of tensorflow code to pytorch code
red line : accuracy of original MNIST imagess of fine-tuned model
blue line : accuracy of adversarial MNIST images of fine-tuned model
- python 3.8+
- pytorch 0.4.1+
- numpy
- tqdm
MIT License
Name | Description |
---|---|
1-layer-linear-classifier | really simple model |
3-layer-linear-classifier | add two layer to 1-layer simple model |
Convnet | simple convolutional model |
This browser does not support PDFs. Please download the PDF to view it: Download PDF.
- VOneNet maybe boosts performance. So we're considering how apply this model to VOneNet