// ⌕ RAG & Vector Retrieval
Contrastive Learning (InfoNCE)
Learns embeddings by pulling positives together and pushing negatives apart.
对比学习与 InfoNCE 2018 Intermediate core
IN PLAIN WORDS
InfoNCE classifies the true pair among many negatives. It is the shared objective behind CLIP, SimCLR, sentence embeddings and retrievers.
Used in: Training embedding and retrieval models
TIMELINE · prerequisites → this → what builds on it
01Original paper
Representation Learning with Contrastive Predictive Coding
Aaron van den Oord, Yazhe Li, Oriol Vinyals · 2018
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Learn first
Contrastive Learning (InfoNCE)
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
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
contrastive-learning)