Canonical: https://knowlpedia.com/real-analysis/critical-point/

# Critical point

A point where the first derivative of a scalar function vanishes

A **critical point** of a differentiable function \(f:U\to \mathbb{R}\) (with \(U\subseteq \mathbb{R}^n\)) is a point \(a\in U\) such that
\[
\nabla f(a)=0,
\]

equivalently, the [Fréchet derivative](https://knowlpedia.com/real-analysis/frechet-derivative/) \(Df(a)\) is the zero linear map.

## Examples

- For \(f(x)=x^3\), the point \(a=0\) is a critical point, but \(0\) is not a local maximum or minimum.
- For \(f(x,y)=x^2+y^2\), the point \((0,0)\) is a critical point and is a (global) minimum.

## Remarks

Critical points are candidates for [local extrema](https://knowlpedia.com/real-analysis/local-extremum/) but need not be extrema. Higher-order information, such as the [Hessian matrix](https://knowlpedia.com/real-analysis/hessian-matrix/), is used to refine classification (see [second derivative tests](https://knowlpedia.com/real-analysis/second-derivative-tests/)).
