Concept模型参数 / Parameters
Parameters
Numerical variables internal to a model that are iteratively adjusted during training to minimize prediction error. In linear regression they appear as slopes and intercepts, while in deep neural networks they can number in the billions.
Timeline (last 90 days)
Oct 13
In deep neural networks, the number of parameters can reach billions, and each parameter participates in the final prediction computation.
Unverified50%