Glossary · Term

end-to-end learning

← all terms

Definition

Training one system from raw input to final answer in one shot, instead of hand-engineering each step in between.

A training paradigm in which a single model is optimized directly on a downstream objective, replacing hand-crafted feature pipelines with learned representations.

Mentioned in 1 episode

  1. 060
    When Splitting One Model Across Three Agents Doubles Its Accuracy