Adding the same scalar-valued function of the input to every class logit leaves all softmax probabilities unchanged. A reference-class constraint removes this common-shift non-identifiability.

Ancestors (9)

  1. Softmax function
  2. Neural network
  3. Statistical learning
  4. Statistical modelling
  5. Statistical model
  6. Probability and statistics
  7. Area of mathematics
  8. Mathematics
  9. Home

Incoming links (2)

View article source

Discussion (0) New discussion

There are no discussions about this article yet.

Articles by others on the same topic (0)

There are currently no matching articles.
See all articles in the same topic Create my own version