Home▸AI & Machine Learning▸3D Backdoor Poisoning and Trigger Reverse-Engineering: Poison Rate, Attack Success and the Anomaly-Index Defence in a Tiny Classifier

🛡️ 3D Backdoor Poisoning and Trigger Reverse-Engineering: Poison Rate, Attack Success and the Anomaly-Index Defence in a Tiny Classifier

A defensive teaching bench built in Blender: a tiny softmax classifier is trained on shape images with a share of poisoned, trigger-stamped samples; watch clean accuracy stay high while the backdoor succeeds, then see the defender recover the trigger mask and flag the class by median-absolute-deviation, checked by finite-difference gradients and a hand-computed anomaly index.

AI & Machine Learning3DAdvanced60 FPS📱 Mobile-adapted🟠 Blender⇄ 2D version
3D Backdoor Poisoning and Trigger Reverse-Engineering: Poison Rate, Attack Success and the Anomaly-Index Defence in a Tiny Classifier ↗ Open standalone
⚙ Under the hood

A defensive teaching bench built in Blender: a tiny softmax classifier is trained on shape images with a share of poisoned, trigger-stamped samples; watch clean accuracy stay high while the backdoor succeeds, then see the defender recover the trigger mask and flag the class by median-absolute-deviation, checked by finite-difference gradients and a hand-computed anomaly index.

Backdoor AttackData PoisoningNeural CleanseTrigger DetectionAnomaly IndexML SecurityDefensive SecurityMachine Learning

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)