Skip to content
Quantoros

Why can reported least-squares cost be half the sum of squared residuals?

MathematicsLevel: FoundationNonlinear least squares

Why can reported least-squares cost be half the sum of squared residuals?

Hints
  • Look for the one-half convention.
  • Constant positive scaling preserves minimizers.

Answer

The objective convention is 0.5Σr_i² for linear loss. The factor simplifies derivatives and does not change the minimizing parameters.

Source and licence: SciPy — nonlinear least squares · BSD-3-Clause · answer: Quantoros editorial

Related questions