Udemy - Adversarial ML - Attack and Defend Neural Networks

  • CategoryOther
  • TypeTutorials
  • LanguageEnglish
  • Total size1.1 GB
  • Uploaded Byfreecoursewb
  • Downloads24
  • Last checkedSep. 27th '26
  • Date uploadedSep. 27th '26
  • Seeders 12
  • Leechers5

Infohash : 0B40C4D1F11C35611A82A386C151C067A45A121D

Adversarial ML: Attack & Defend Neural Networks

https://WebToolTip.com

Published 9/2026
Created by Bayt Al Hikmah
MP4 | Video: h264, 1280x720 | Audio: AAC, 44.1 KHz, 2 Ch
Level: All Levels | Genre: eLearning | Language: English | Duration: 112 Lectures ( 25h 24m ) | Size: 1.2 GB

What you'll learn
âš¡ Build a full PyTorch training and evaluation pipeline with reproducibility discipline
âš¡ Implement the core white-box attack arsenal from scratch
âš¡ Simulate realistic black-box threats against a live FastAPI inference service
âš¡ Attack beyond evasion
âš¡ Engineer real defenses and measure them honestly
âš¡ Build ML governance evidence
âš¡ Automate MLOps and supply-chain security
âš¡ Deploy and operate production inference safely
âš¡ Design for sovereignty
âš¡ Deliver a capstone-grade Sovereign Adversarial ML Defense Platform

Requirements
❗ Knowledge: Basic Python (functions, running scripts) and basic terminal comfort. No prior deep learning, PyTorch, or adversarial ML experience required — Module 1 trains your first neural network from a bare workstation up. Basic familiarity with machine learning concepts (what training and accuracy mean) helps but isn't required — every concept is explained before it's used. No prior Kubernetes, MLOps, or DevSecOps experience needed — Modules 8–9 build those skills from scratch. Software (all free/open-source): Python 3.11+, Git, Docker. Python packages installed via pip in Lab 1: PyTorch, torchvision, scikit-learn, FastAPI, Uvicorn, pytest, MLflow, the Adversarial Robustness Toolbox (ART), Foolbox, Evidently, Great Expectations, Prometheus client, OpenTelemetry — all free and open-source. kubectl (optional, for Module 9's Kubernetes labs) — a local cluster, not a cloud account. Optional: Trivy and Cosign for supply-chain labs — the course provides graceful fallbacks if they aren't installed. Hardware: CPU-only is fully sufficient — this course intentionally uses a small, fast scikit-learn digits dataset (8x8 grayscale images) so every lab runs in seconds without a GPU. 5GB+ free disk space, 4GB+ RAM. No cloud account, no GPU, and no real production data required — every lab uses a safe, built-in benchmark dataset with no licensing or download concerns.

Description
This course contains the use of artificial intelligence.

Files:

[ WebToolTip.com ] Udemy - Adversarial ML - Attack and Defend Neural Networks
  • Get Bonus Downloads Here.url (0.2 KB)
    • ~Get Your Files Here ! 1 - Introduction
    • 1. Introduction.mp4 (115.1 MB)
    10 - Module 9 Production Serving, Observability, and Reliability
    • 100. Lab 90 Deploy to a Local Kubernetes Cluster.html (41.3 KB)
    • 101. Lab 91 Add Prometheus Metrics to the API.html (46.9 KB)
    • 102. Lab 92 Add OpenTelemetry Tracing.html (37.1 KB)
    • 103. Lab 93 Define Service Level Objectives.html (52.4 KB)
    • 104. Lab 94 Run a Local Load Test.html (39.8 KB)
    • 105. Lab 95 Implement Canary Deployment Manifests.html (38.5 KB)
    • 106. Lab 96 Production Incident and Rollback Drill.html (51.6 KB)
    • 98. Production Serving, Observability, and Reliability.mp4 (93.0 MB)
    • 99. Lab 89 Create Kubernetes Manifests for Inference.html (40.8 KB)
    11 - Module 10 Sovereign Deployment, Decentralized Research, and Capstone
    • 107. Sovereign Deployment, Decentralized Research, and Capstone.mp4 (91.8 MB)
    • 108. Lab 97 Build a Local-First Sovereign Deployment Profile.html (46.5 KB)
    • 109. Lab 98 Create a Decentralized Research Reproducibility Pack.html (47.7 KB)
    • 110. Lab 99 Capstone Architecture Design Review.html (46.3 KB)
    • 111. Lab 100 Sovereign Adversarial ML Defense Capstone.html (50.4 KB)
    12 - Conclusion
    • 112. Conclusion.mp4 (102.0 MB)
    2 - Module 1 Foundations, Local Environment, and Initial Success
    • 10. Lab 8 Add a Minimal Defense by Input Clipping and Noise.html (36.0 KB)
    • 11. Lab 9 Convert the First Experiment into a Test.html (34.4 KB)
    • 12. Lab 10 Initial Success Milestone Report.html (35.0 KB)
    • 2. Foundations, Local Environment, and Initial Success.mp4 (72.4 MB)
    • 3. Lab 1 Build the Reproducible Adversarial ML Workstation.html (45.8 KB)
    • 4. Lab 2 Create the Course Repository Contract.html (37.6 KB)
    • 5. Lab 3 Load a Safe Benchmark Dataset.html (35.2 KB)
    • 6. Lab 4 Train the First Neural Network Baseline.html (35.6 KB)
    • 7. Lab 5 Build the First Evaluation Script.html (31.3 KB)
    • 8. Lab 6 Produce the First FGSM Attack.html (32.9 KB)
    • 9. Lab 7 Visualize Clean vs Adversarial Inputs.html (33.8 KB)
    3 - Module 2 Neural Network Baselines and Evaluation Discipline
    • 13. Neural Network Baselines and Evaluation Discipline.mp4 (97.2 MB)
    • 14. Lab 11 Build a Config-Driven Training Script.html (39.7 KB)
    • 15. Lab 12 Add Deterministic Seeds and Reproducibility Checks.html (38.8 KB)
    • 16. Lab 13 Add MLflow Experiment Tracking.html (36.6 KB)
    • 17. Lab 14 Build a Model Factory.html (36.4 KB)
    • 18. Lab 15 Create a Metric Registry.html (34.7 KB)
    • 19. Lab 16 Build Dataset Integrity Hashing.html (42.4 KB)
    • 20. Lab 17 Add Dataset Schema Validation.html (48.1 KB)
    • 21. Lab 18 Create a Confusion Matrix Report.html (45.0 KB)
    • 22. Lab 19 Build a Clean Accuracy Gate.html (34.4 KB)
    • 23. Lab 20 Baseline Readiness Review.html (37.0 KB)
    4 - Module 3 White-Box Evasion Attacks
    • 24. White-Box Evasion Attacks.mp4 (103.3 MB)
    • 25. Lab 21 Implement FGSM as a Reusable Attack Function.html (38.3 KB)
    • 26. Lab 22 Run an Epsilon Sweep.html (33.1 KB)
    • 27. Lab 23 Implement Targeted FGSM.html (34.7 KB)
    • 28. Lab 24 Implement Basic Iterative Method.html (38.3 KB)
    • 29. Lab 25 Implement Projected Gradient Descent.html (39.0 KB)
    • 30. Lab 26 Add Carlini-Wagner Style Margin Loss.html (46.4 KB)
    • 31. Lab 27 Evaluate Transfer Across Model Architectures.html (41.9 KB)
    • 32. Lab 28 Use Foolbox for Independent Attack Validation.html (36.7 KB)
    • 33. Lab 29 Use ART for Attack Validation.html (34.7 KB)
    • 34. Lab 30 Build Attack Comparison Dashboard Data.html (46.2 KB)
    • 35. Lab 31 Create a Robustness Curve Plot.html (31.9 KB)
    • 36. Lab 32 White-Box Attack Review Gate.html (43.9 KB)
    5 - Module 4 Black-Box, Transfer, and Query-Based Attacks
    • 37. Black-Box, Transfer, and Query-Based Attacks.mp4 (101.3 MB)
    • 38. Lab 33 Wrap the Model Behind a FastAPI Inference Service.html (33.2 KB)
    • 39. Lab 34 Build a Query Client.html (36.1 KB)
    • 40. Lab 35 Simulate Score-Based Black-Box Probing.html (41.8 KB)
    • 41. Lab 36 Simulate Decision-Only Boundary Search.html (39.8 KB)
    • 42. Lab 37 Add Query Budget Accounting.html (39.2 KB)
    • 43. Lab 38 Add API Rate Limiting.html (35.9 KB)
    • 44. Lab 39 Evaluate Transfer Attack Against API-Only Access.html (38.4 KB)
    • 45. Lab 40 Build an Abuse Detection Log.html (47.4 KB)
    • 46. Lab 41 Detect Suspicious Query Bursts.html (37.1 KB)
    • 47. Lab 42 Black-Box Threat Review.html (46.6 KB)
    6 - Module 5 Poisoning, Backdoors, Extraction, and Privacy Attacks
    • 48. Poisoning, Backdoors, Extraction, and Privacy Attacks.mp4 (88.1 MB)
    • 49. Lab 43 Simulate Label-Flipping Poisoning.html (39.0 KB)
    • 50. Lab 44 Train on Poisoned Data and Compare.html (40.9 KB)
    • 51. Lab 45 Build a Simple Backdoor Trigger.html (50.5 KB)
    • 52. Lab 46 Train and Evaluate Backdoor Success.html (42.7 KB)
    • 53. Lab 47 Detect Backdoor Candidate Samples by Pixel Statistics.html (31.2 KB)
    • 54. Lab 48 Simulate Model Extraction with API Queries.html (35.8 KB)
    • 55. Lab 49 Train an Extracted Surrogate Model.html (43.6 KB)
    • 56. Lab 50 Add Output Minimization Defense.html (38.0 KB)
    • 57. Lab 51 Run Membership Inference Risk Simulation.html (36.3 KB)
    • 58. Lab 52 Add Privacy-Aware Logging Controls.html (35.9 KB)
    • 59. Lab 53 Map Threats to MITRE ATLAS-Style Records.html (43.4 KB)
    • 60. Lab 54 Create a Model Abuse Case Register.html (49.9 KB)
    • 61. Lab 55 Training-Time Threat Review Gate.html (40.8 KB)
    7 - Module 6 Defensive Engineering and Robust Training
    • 62. Defensive Engineering and Robust Training.mp4 (76.6 MB)
    • 63. Lab 56 Train with FGSM Adversarial Augmentation.html (31.7 KB)
    • 64. Lab 57 Evaluate Robust Model Against PGD.html (28.0 KB)
    • 65. Lab 58 Implement PGD Adversarial Training.html (36.0 KB)
    • 66. Lab 59 Compare Clean-Robust Tradeoffs.html (41.0 KB)
    • 67. Lab 60 Add Confidence-Based Rejection.html (44.4 KB)
    • 68. Lab 61 Build Entropy-Based Detection.html (36.9 KB)

    • Code:

      • udp://coeus.torrentonline.cc:42069/announce
      • https://edge-team.cc/announce
      • https://tracker.madtia.cc/announce
      • udp://tracker.1h.is:1337/announce
      • udp://tracker.t-1.org:6969/announce
      • udp://open.stealth.si:80/announce
      • udp://whybother.torrentonline.cc:42069/announce
      • udp://obey.torrentonline.cc:42069/announce
      • udp://archive.torrentonline.cc:42069/announce
      • https://tracker.7471.top:443/announce
      • https://tracker.pmman.tech:443/announce
      • https://torrents.tmtime.dev:443/announce
      • http://tracker.moeblog.cn:443/announce
      • http://tracker.lilithraws.org:443/announce
      • http://tr.highstar.shop:80/announce