Autoencoder for anomaly detection — how?

medium

Answer

$Train AE on 'normal' data — it learns to reconstruct normal patterns well.$
$At inference, compute reconstruction error ||x - x̂||²; high error = anomaly.$
$Works well when normal data has low-dimensional manifold structure (images, time series).$
$Pitfalls: (1) if AE is too powerful, it reconstructs anomalies too — use small bottleneck; (2) contamination in training data \to semi-supervised or filter first.$

Check yourself — multiple choice

Same as SVM
Train AE on 'normal' data; ||x - x̂||² > threshold = anomaly; works for images / time series; too-powerful AE also reconstructs anomalies → small bottleneck
Random
Only supervised

AE anomaly: high reconstruction error = anomaly; watch for over-capacity.

#anomaly-detection#deep-learning

214 interview questions in this topic.