Deep learning
Deep learning uses layered parameterized functions to learn representations at multiple levels of abstraction. Depth is an architectural property; useful performance still depends on training data, objectives, optimization, and compute.
Overview. Deep learning uses layered parameterized functions to learn representations at multiple levels of abstraction.
Context. Depth is an architectural property; useful performance still depends on training data, objectives, optimization, and compute.
Claims and evidence
Claim ledger, 2Deep learning uses layered parameterized functions to learn representations at multiple levels of abstraction.
Depth is an architectural property; useful performance still depends on training data, objectives, optimization, and compute.