A→Z
A2ZAI
Back to Glossary
concepts

Embedding

Dense vector representation of data that captures semantic meaning.

Share:

Definition

Embeddings convert discrete data (words, sentences, images) into continuous vector representations in a high-dimensional space.

Key Properties: - Similar items have similar vectors - Enable mathematical operations on meaning - Typically 384-4096 dimensions

  • **Types:**
  • Word embeddings: Word2Vec, GloVe
  • Sentence embeddings: SBERT, Ada-002
  • Image embeddings: CLIP, ResNet features
  • Code embeddings: CodeBERT

Applications: - Semantic search - Recommendation systems - Clustering and classification - RAG systems

Popular Models: - OpenAI text-embedding-ada-002 - Cohere embed - Voyage AI - BGE, E5

Examples

Converting "king" and "queen" to vectors where their difference equals the "male-female" direction.

Want more AI knowledge?

Get bite-sized AI concepts delivered to your inbox.

Free daily digest. No spam, unsubscribe anytime.

Discussion