// ⬡ Deep Learning Foundations
Dropout
Randomly drops neurons during training to prevent overfitting.
2012 Beginner core
IN PLAIN WORDS
Each step zeroes random units so the network cannot rely on any single one — like averaging many sub-networks.
Used in: CNNs, MLPs, some Transformers
01Original paper
Improving neural networks by preventing co-adaptation of feature detectors
Geoffrey E. Hinton, Nitish Srivastava, Alex Krizhevsky, et al. · 2012
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Related:Batch Normalization
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
04Where the course teaches it
On learning paths:Zero to Deep Learning
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
dropout)