Deep learning

Branch of AI based on multi-layer artificial neural networks.

📖 Full definition

Deep learning uses neural networks organised in successive layers to learn hierarchical data representations. In face recognition, the first layers detect edges, the next ones shapes, the last ones identities. The more layers a model has, the more subtleties it can capture — at the cost of higher training and inference cost.

💡 Concrete example

ArcFace, which Look Like Me draws inspiration from, is a deep learning network trained on millions of faces to produce discriminative embeddings.

🔄 Synonyms

machine learning

🔗 See also

🪞 Test resemblance with our AI

Start my free test →