Adversarial Examples and Robust Evaluation: From FGSM to a scikit-learn Digits Experiment
AI Learning Project
Adversarial Examples and Robust Evaluation: From FGSM to a scikit-learn Digits Experiment
A defensive robustness evaluation walkthrough using the FGSM equation and a safe scikit-learn digits toy experiment.
AI Learning Project

Adversarial Examples and Robust Evaluation: From FGSM to a scikit-learn Digits Experiment

Reading info

Level: Professional Reading time: 11 min
  • Adversarial Examples
  • FGSM
  • Robust Evaluation
  • scikit-learn
Open knowledge map

Adversarial examples are not arbitrary noise distributions; they are mathematically precise vectors computed by optimizing an adversarial objective function over a neural network’s loss manifold. A production-grade evaluation cannot rely on clean accuracy metrics. It demands rigorous assessment of empirical robustness against bounded perturbations, analysis of the Jacobian matrix, and computation of the defense’s systemic latency and accuracy trade-offs.

This article deconstructs the mathematical framework of gradient-based attacks (FGSM and PGD), provides PyTorch implementations for Red Teams, and details the production pipeline architectures required for adversarial defense.

1. The Mathematical Boundaries of Threat Models

An adversarial evaluation is mathematically meaningless without defining the feasible set of the attacker. The threat model is parameterized by:

  • Attacker Knowledge: White-box (full access to ( theta ), architectures, and gradients ( nabla_x J )) vs. Black-box (zero-th order optimization via queries).
  • Perturbation Constraint (( L_p ) Norm): The perturbation ( delta ) is bounded by ( |delta|_p le epsilon ). Common norms include ( L_infty ) (maximum pixel change) and ( L_2 ) (Euclidean distance).
  • Objective Function: Untargeted (( argmax_delta J(theta, x+delta, y) )) vs. Targeted (( argmin_delta J(theta, x+delta, y_{target}) )).

2. Fast Gradient Sign Method (FGSM)

FGSM is a single-step gradient-based attack that linearizes the loss function ( J ) around the input ( x ). Utilizing a first-order Taylor expansion, the attacker maximizes the loss under an ( L_infty ) constraint.

The mathematical formulation is:

[ delta = epsilon cdot text{sign}(nabla_x J(theta, x, y)) ]

[ x_{adv} = text{clip}(x + delta, x_{min}, x_{max}) ]

PyTorch Implementation of FGSM

import torch
import torch.nn as nn

def fgsm_attack(model, images, labels, epsilon, criterion):
    images.requires_grad = True
    outputs = model(images)
    loss = criterion(outputs, labels)
    
    # Compute Jacobian / Gradients wrt input
    model.zero_grad()
    loss.backward()
    data_grad = images.grad.data
    
    # Create perturbation
    sign_data_grad = data_grad.sign()
    perturbed_images = images + epsilon * sign_data_grad
    
    # Project back to valid input domain (e.g., [0, 1])
    perturbed_images = torch.clamp(perturbed_images, 0, 1)
    return perturbed_images

3. Projected Gradient Descent (PGD)

While FGSM is computationally efficient, it underfits the adversarial objective. Projected Gradient Descent (PGD) is the universal first-order adversary. It solves the constrained optimization problem via iterative gradient steps, projecting the perturbation back onto the ( epsilon )-ball after each step.

The update rule for step ( t+1 ) is:

[ x^{t+1} = Pi_{x+mathcal{S}} left( x^t + alpha cdot text{sign}(nabla_x J(theta, x^t, y)) right) ]

Where ( alpha ) is the step size and ( Pi_{x+mathcal{S}} ) is the projection operator onto the ( L_p ) ball.

PyTorch Implementation of PGD

def pgd_attack(model, images, labels, epsilon, alpha, iters, criterion):
    perturbed_images = images.clone().detach()
    # Random start within epsilon ball
    perturbed_images = perturbed_images + torch.empty_like(perturbed_images).uniform_(-epsilon, epsilon)
    perturbed_images = torch.clamp(perturbed_images, 0, 1)
    
    for _ in range(iters):
        perturbed_images.requires_grad = True
        outputs = model(perturbed_images)
        loss = criterion(outputs, labels)
        
        model.zero_grad()
        loss.backward()
        
        with torch.no_grad():
            adv_images = perturbed_images + alpha * perturbed_images.grad.sign()
            eta = torch.clamp(adv_images - images, min=-epsilon, max=epsilon)
            perturbed_images = torch.clamp(images + eta, 0, 1)
            
    return perturbed_images

4. Red/Blue Team Post-Mortem: Production Architecture Defenses

In production pipelines, basic “random noise” defenses are completely defeated by Expectation Over Transformation (EOT). Real-world mitigation relies on architectural integration:

  • Adversarial Training Logic: The empirical risk minimization is modified to a min-max saddle point problem:

    [ min_theta mathbb{E}_{(x,y)sim mathcal{D}} left[ max_{|delta|_p le epsilon} J(theta, x+delta, y) right] ]
    Models are continuously trained on PGD-generated samples. This lowers the curvature of the loss surface but comes at the cost of the “accuracy-robustness trade-off” (diminished clean accuracy).
  • Gradient Masking & Obfuscation (A Warning): Blue teams often inadvertently introduce shattered gradients (e.g., non-differentiable preprocessing). Red teams bypass this using Backward Pass Differentiable Approximation (BPDA). True defense requires verifying robustness via black-box transfer attacks.
  • Inference Abstention & Out-of-Distribution (OOD) Detection: Deploying Mahalanobis distance metrics on deep feature representations to detect inputs lying far from the clean training manifold.

5. Robust Evaluation Reporting Standards

A production security audit must yield an evaluation matrix:

  • Clean Accuracy vs. PGD-100 (100 iterations) Accuracy across a spectrum of ( epsilon ) budgets.
  • Evaluation of gradient-free attacks (e.g., SPSA) to certify that defenses are not merely relying on gradient obfuscation.
  • System latency overhead introduced by dynamic OOD detection modules.

6. Robustness Audit Matrix

The strongest adversarial evaluation reports include both attack strength and defense side effects. A model should not be called robust unless the evaluation records the attack budget, adaptive checks, and production impact.

Audit dimension Required measurement Interpretation Red flag
Attack budget ( epsilon ), norm type, PGD steps, step size, random restarts Defines what the adversary is actually allowed to do Only reporting one weak FGSM result and claiming broad robustness
Adaptive attack BPDA/EOT or gradient-free transfer checks when preprocessing is non-differentiable Separates real robustness from gradient masking Robust accuracy is high for white-box gradients but low for black-box transfer
Clean accuracy trade-off Clean, FGSM, PGD-20, PGD-100, and OOD accuracy in the same report Shows whether the defense is useful for normal traffic Robustness improves only by making the model reject or misclassify clean data
Runtime cost Median and p95 latency with OOD detection or input purification enabled Connects security controls to deployability Defense requires many forward passes and cannot meet service latency budgets

7. References

Internal links

Related tutorials

Review foundation AI Security Threat Modeling

Build a defense map with NIST adversarial ML, MITRE ATLAS, and OWASP LLM risks.

 Continue next Data Poisoning and Backdoor Defense

Study poison rate, trigger behavior, attack success rate, and training pipeline controls.

 Learning path AI Learning Path

Read AI foundations, machine learning, CNN work, traffic defense, and AI security in order.

 Resources Resource Library

Download companion code, datasets, diagrams, model files, and notes.

 Topic hub Student AI Projects

Connect handwritten digits, browser demos, and CNN projects into a portfolio path.

Search questions

FAQ

Who is this article for?

This article is for readers who want a professional-level guide to Adversarial Examples and Robust Evaluation. It takes about 11 min and focuses on Adversarial Examples, FGSM, Robust Evaluation, scikit-learn.

What should I read next?

The recommended next step is Data Poisoning and Backdoor Defense, so the article connects into a longer learning route instead of ending as an isolated note.

Does this article include runnable code or companion resources?

Yes. Use the run notes, resource cards, and download links on the page to reproduce the example or inspect the companion files.

How does this article fit into the larger site?

It is connected to the article context block, learning routes, resources, and project timeline so readers can move from concept to implementation.

Article context

AI Learning Project

A practical route from AI concepts to machine learning workflow, evaluation, neural networks, Python practice, handwritten digits, a CIFAR-10 CNN, adversarial traffic-defense notes, and AI security.

Level: Professional Reading time: 11 min
  • Adversarial Examples
  • FGSM
  • Robust Evaluation
  • scikit-learn

Your next step

Continue: Data Poisoning and Backdoor Defense
Review the foundation Open resource
Other language version 对抗样本与鲁棒评估：从 FGSM 公式到 scikit-learn 数字分类实验
Share summary Adversarial Examples and Robust Evaluation

Evaluate clean and perturbed accuracy with an FGSM-style digits experiment.

Download share card Open share center

Companion resources

Leave a Reply

Project timeline

Published posts

  1. AI Basics Learning Roadmap Separate AI, machine learning, and deep learning before going into implementation details.
  2. Machine Learning Workflow Follow the practical path from data and features to training, prediction, and evaluation.
  3. Model Training and Evaluation Understand loss, overfitting, train/test splits, accuracy, recall, and F1.
  4. Neural Network Basics Move from perceptrons to activation, forward propagation, backpropagation, and training loops.
  5. Matrix Calculus for Neural Networks Derive dL/dW for y = Wx + b and verify it with finite differences.
  6. Backpropagation as a Computation Graph Trace local gradients through ReLU and softmax cross-entropy in a two-layer MLP.
  7. Gradient Descent and Optimizer Geometry Compare gradient descent, momentum, and Adam on a visible quadratic loss surface.
  8. Convolution and Receptive Field Math Compute convolution output size, receptive fields, channel mixing, and im2col layout.
  9. Transformer Attention Math Hand-calculate Q/K/V scores, softmax weights, masks, multi-head structure, and KV cache.
  10. Python AI Mini Practice Run a small scikit-learn classification task and read the experiment output.
  11. Handwritten Digit Dataset Basics Read train.csv, test.csv, labels, and the flattened 28 by 28 pixel layout before training the classifier.
  12. Handwritten Digit Softmax in C Follow the C implementation from logits and softmax probabilities to confusion matrices and submission export.
  13. Handwritten Digit Playground Notes See how the offline classifier was adapted into a browser demo with drawing input and probability output.
  14. CIFAR-10 Tiny CNN Tutorial in C Build and train a small convolutional neural network for CIFAR-10 image classification, then read its loss and accuracy output.
  15. High-Entropy Traffic Defense Notes Study encrypted metadata leaks, entropy, traffic classifiers, and a defensive Python chaffing prototype.
  16. AI Security Threat Modeling Build a defense map with NIST adversarial ML, MITRE ATLAS, and OWASP LLM risks.
  17. Adversarial Examples and Robust Evaluation Evaluate clean and perturbed accuracy with an FGSM-style digits experiment.
  18. Data Poisoning and Backdoor Defense Study poison rate, trigger behavior, attack success rate, and training pipeline controls.
  19. Model Privacy and Extraction Defense Measure membership inference signal and surrogate fidelity against a local toy model.
  20. LLM, RAG, and Agent Security Separate instructions from data and enforce tool permissions against indirect prompt injection.

Published resources

  1. Python AI practice code guide The article includes a runnable scikit-learn classification script.
  2. digit_softmax_classifier.c The C source for the handwritten digit softmax classifier.
  3. train.csv.zip Compressed handwritten digit training set with 42000 labeled samples.
  4. test.csv.zip Compressed handwritten digit test set with 28000 unlabeled samples.
  5. sample_submission.csv The official submission format example for checking the final output columns.
  6. submission.csv The prediction file generated by the current C project.
  7. digit-playground-model.json The compact softmax demo model and sample set used by the browser playground.
  8. digit-sample-grid.svg A small handwritten digit preview grid extracted from the training set.
  9. Handwritten digit project bundle Contains the source file, compressed datasets, submission files, browser model, and preview grid.
  10. cifar10_tiny_cnn.c source Single-file C tiny CNN with CIFAR-10 loading, convolution, pooling, softmax, and backpropagation.
  11. model_weights.bin sample weights Model weights generated by one local small-sample run.
  12. test_predictions.csv sample predictions Sample test prediction output from the CIFAR-10 tiny CNN.
  13. CNN project explanation PDF Companion explanation material for the CNN project.
  14. Virtual Mirror redacted code skeleton A redacted mld_chaffing_v2.py control-flow skeleton with secrets, node topology, and target lists removed.
  15. Virtual Mirror stress-test template A redacted CSV template for CPU, memory, peak threads, pulse rate, latency, and error measurements.
  16. Virtual Mirror classifier-evaluation template A CSV template for TP, FN, FP, TN, accuracy, precision, recall, F1, ROC-AUC, entropy, and JS divergence.
  17. Virtual Mirror resource notes Notes explaining why the public resources include only redacted code, test templates, and architecture context.
  18. AI Security Lab README Setup, safety boundaries, and quick-run commands for the AI Security series.
  19. AI Security Lab full bundle Includes safe toy scripts, result CSVs, risk register, attack-defense matrix, and architecture diagram.
  20. AI security risk register CSV risk register template for AI threat modeling and release review.
  21. AI attack-defense matrix Maps attack surface, toy demo, metric, and defensive control into one CSV table.
  22. AI Security Lab architecture diagram Shows threat modeling, robustness, data integrity, model privacy, and RAG guardrails.
  23. FGSM digits robustness script FGSM-style perturbation and accuracy-drop experiment for a local digits classifier.
  24. Data poisoning and backdoor toy script Demonstrates poison rate, trigger behavior, and attack success rate on digits.
  25. Model privacy and extraction toy script Outputs membership AUC, target accuracy, surrogate fidelity, and surrogate accuracy.
  26. RAG prompt injection guard toy script Uses a deterministic toy agent to demonstrate external-data demotion and tool-policy blocking.
  27. Deep Learning Math Lab README Setup commands, script entry points, generated outputs, and figure notes for the math series.
  28. Deep learning math full lab bundle Bundles NumPy scripts, CSV outputs, formula diagrams, loss contours, convolution figures, and attention heatmaps.
  29. Gradient check results CSV Stores MSE analytic gradients, finite-difference gradients, and error norms.
  30. Optimizer path CSV Step-by-step coordinates and loss for gradient descent, momentum, and Adam on a 2D quadratic.
  31. Attention weights CSV Scores, softmax weights, and context vectors for a three-token scaled dot-product attention example.
  32. Deep learning math figure set Includes matrix shapes, computation graphs, loss contours, convolution scans, and attention heatmaps.
  33. Deep learning math interactive visualizer Browser modules for gradient checking, optimizer paths, convolution output size, and attention heatmaps.
  34. Deep Learning topic share card A 1200x630 SVG card for sharing the Deep Learning / CNN topic hub.
  35. Machine Learning From Scratch share card A 1200x630 SVG card for the K-means, Iris, and ML workflow topic hub.
  36. Student AI Projects share card A 1200x630 SVG card for handwritten digits, C classifiers, and browser demos.
  37. CNN convolution scan animation An 8-second Remotion animation showing how a 3x3 convolution kernel scans an input and builds a feature map.

Current route

  1. AI Basics Learning Roadmap Learning path step
  2. Machine Learning Workflow Learning path step
  3. Model Training and Evaluation Learning path step
  4. Neural Network Basics Learning path step
  5. Matrix Calculus for Neural Networks Learning path step
  6. Backpropagation as a Computation Graph Learning path step
  7. Gradient Descent and Optimizer Geometry Learning path step
  8. Convolution and Receptive Field Math Learning path step
  9. Transformer Attention Math Learning path step
  10. LLM Visualizer Learning path step
  11. Python AI Mini Practice Learning path step
  12. Handwritten Digit Dataset Basics Learning path step
  13. Handwritten Digit Softmax in C Learning path step
  14. Handwritten Digit Playground Notes Learning path step
  15. CIFAR-10 Tiny CNN Tutorial in C Learning path step
  16. High-Entropy Traffic Defense Notes Learning path step
  17. AI Security Threat Modeling Learning path step
  18. Adversarial Examples and Robust Evaluation Learning path step
  19. Data Poisoning and Backdoor Defense Learning path step
  20. Model Privacy and Extraction Defense Learning path step
  21. LLM, RAG, and Agent Security Learning path step

Next notes

  1. Add more image-classification and error-analysis cases
  2. Turn common metrics into a quick reference
  3. Add more AI security defense experiment notes
Scroll down