README.md

---
license: mit
library_name: transformers
datasets:
- uoft-cs/cifar10
pipeline_tag: image-classification
tags:
- computer-vision
- cnn
- cifar10
- adversarial-robustness
- downsampling
- anti-aliasing
metrics:
- accuracy
---

## NULA 

Anti-aliased residual CNN for CIFAR-10 image classification designed to improve robustness against perturbations that exploit downsampling.

# Problem

Downsampling operations are linear maps from a high-dimensional space to a lower-dimensional one.

By the Rank-Nullity theorem, this matrix has a massive NULL space. 

An attacker can exploit this: they utilize the discarded samples of these downsampling operations as extra degrees of freedom. 

![downsampling null space](https://visionbook.mit.edu/figures/upsamplig_downsampling/toy_example_book2.png)https://visionbook.mit.edu/figures/upsamplig_downsampling/toy_example_book2.png)
*Source: MIT Vision Book*

By sculpting perturbations with components in the null space of the downsampling operator, they spread energy across frequencies that are discarded during striding.

The result is an image perceptually identical to the original, with a manipulated activation pattern.

```python

```
## Approach

[BlurPool: what it does mechanically and why it addresses the problem]
[SE blocks: what they add]

## Architecture

[stem → s1 → s2 → s3 → head, with channel dims]
[BlurPool replaces strided conv — note this explicitly]

## Training Statistics 

*first evaluation* on CIFAR-10 test set (10,000 images). All perturbations applied at inference
the model never saw test distribution during training.

| Perturbation | Accuracy | Drop |
|---|---|---|
| Clean | 91.95% | — |
| Resize ×0.5 (bilinear) | 59.83% | −32.12% |
| Resize ×0.25 (bilinear) | 24.82% | −67.13% |
| Decimate ×2 | 30.03% | −61.92% |
| Checkerboard ε=0.03 | 75.47% | −16.48% |
| Checkerboard ε=0.05 | 44.99% | −46.96% |

*second evaluation* on the CIFAR-10 test set (10,000 images). Perturbations applied at inference only.

The adversarially trained variant was trained from scratch under a modified data distribution. 

| Perturbation | Accuracy | Δ from 1st eval |
|---|---|---|
| Clean | 89.42% | −2.53% |
| Resize ×0.5 | 85.37% | +25.54% |
| Resize ×0.25 | 71.80% | +46.98% |
| Decimate ×2 | 85.02% | +54.99% |
| Checkerboard ε=0.03 | 89.43% | +13.96% |
| Checkerboard ε=0.05 | 89.39% | +44.40% |

The adversarial variant trades 2.53% clean accuracy for substantial robustness across all tested perturbations.
The checkerboard attack —a direct null-space exploit against stride-2 downsampling — drops from 44.99% to near-clean 89.39%.

## Usage

NULA is hosted on the HuggingFace Hub and can be loaded directly via the transformers library. 
```python
import torch as pt
from transformers import AutoModelForImageClassification

model = AutoModelForImageClassification.from_pretrained(
    "mamapearli/NULA-base-cifar10-v0",
    trust_remote_code=True
)

image = pt.randn(1, 3, 32, 32)
with pt.no_grad():
    output = model(pixel_values=image)

predicted_class = output.logits.argmax(dim=-1).item()
print(model.config.id2label[predicted_class])
```
Input tensors should be shape (B, C, H, W).

## Citation

If you use this model or repository in your research, please cite:

```bibtex
@misc{mamapearl_nula_2026,
  author = {MamaPearl},
  title = {NULA: Anti-Aliased SE-CNN for CIFAR-10},
  year = {2026},
  publisher = {Hugging Face},
  url = {https://huggingface.co/MamaPearl/nula-base-cifar10-v0}
}
```

# Authors

* **MamaPearl** ([@MamaPearl](https://github.com))  
  *Main Contributor* 

## Contact & Socials
| Platform | Link |
| :--- | :--- |
| **GitHub** | [github.com/MamaPearl](https://github.com) |
| **Instagram** | [@mamapearli](https://www.instagram.com/mamapearli/) | 
## License

This project is licensed under the MIT License. See [LICENSE](LICENSE) for more information.