Migdal et al, Equation (51)

Percentage Accurate: 99.4% → 99.4%
Time: 10.8s
Alternatives: 11
Speedup: 0.6×

Specification

?
\[\begin{array}{l} \\ \frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \end{array} \]
(FPCore (k n)
 :precision binary64
 (* (/ 1.0 (sqrt k)) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0))))
\begin{array}{l}

\\
\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}
\end{array}

Sampling outcomes in binary64 precision:

Local Percentage Accuracy vs ?

The average percentage accuracy by input value. Horizontal axis shows value of an input variable; the variable is choosen in the title. Vertical axis is accuracy; higher is better. Red represent the original program, while blue represents Herbie's suggestion. These can be toggled with buttons below the plot. The line is an average while dots represent individual samples.

Accuracy vs Speed?

Herbie found 11 alternatives:

AlternativeAccuracySpeedup
The accuracy (vertical axis) and speed (horizontal axis) of each alternatives. Up and to the right is better. The red square shows the initial program, and each blue circle shows an alternative.The line shows the best available speed-accuracy tradeoffs.

Initial Program: 99.4% accurate, 1.0× speedup?

\[\begin{array}{l} \\ \frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \end{array} \]
(FPCore (k n)
 :precision binary64
 (* (/ 1.0 (sqrt k)) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0))))
\begin{array}{l}

\\
\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}
\end{array}

Alternative 1: 99.4% accurate, 0.9× speedup?

\[\begin{array}{l} \\ \begin{array}{l} t_0 := \mathsf{PI}\left(\right) \cdot \left(n \cdot 2\right)\\ \frac{\sqrt{t\_0}}{{t\_0}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \end{array} \end{array} \]
(FPCore (k n)
 :precision binary64
 (let* ((t_0 (* (PI) (* n 2.0))))
   (/ (sqrt t_0) (* (pow t_0 (/ k 2.0)) (sqrt k)))))
\begin{array}{l}

\\
\begin{array}{l}
t_0 := \mathsf{PI}\left(\right) \cdot \left(n \cdot 2\right)\\
\frac{\sqrt{t\_0}}{{t\_0}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}}
\end{array}
\end{array}
Derivation
  1. Initial program 99.5%

    \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
  2. Add Preprocessing
  3. Step-by-step derivation
    1. lift-*.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\color{blue}{\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}}^{\left(\frac{1 - k}{2}\right)} \]
    2. *-commutativeN/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\color{blue}{\left(n \cdot \left(2 \cdot \mathsf{PI}\left(\right)\right)\right)}}^{\left(\frac{1 - k}{2}\right)} \]
    3. lift-*.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(n \cdot \color{blue}{\left(2 \cdot \mathsf{PI}\left(\right)\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \]
    4. associate-*r*N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\color{blue}{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}}^{\left(\frac{1 - k}{2}\right)} \]
    5. lift-PI.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(\left(n \cdot 2\right) \cdot \color{blue}{\mathsf{PI}\left(\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \]
    6. add-sqr-sqrtN/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(\left(n \cdot 2\right) \cdot \color{blue}{\left(\sqrt{\mathsf{PI}\left(\right)} \cdot \sqrt{\mathsf{PI}\left(\right)}\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \]
    7. associate-*r*N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\color{blue}{\left(\left(\left(n \cdot 2\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right)}}^{\left(\frac{1 - k}{2}\right)} \]
    8. lower-*.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\color{blue}{\left(\left(\left(n \cdot 2\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right)}}^{\left(\frac{1 - k}{2}\right)} \]
    9. lower-*.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(\color{blue}{\left(\left(n \cdot 2\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right)} \cdot \sqrt{\mathsf{PI}\left(\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \]
    10. lower-*.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(\left(\color{blue}{\left(n \cdot 2\right)} \cdot \sqrt{\mathsf{PI}\left(\right)}\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \]
    11. lift-PI.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(\left(\left(n \cdot 2\right) \cdot \sqrt{\color{blue}{\mathsf{PI}\left(\right)}}\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \]
    12. lower-sqrt.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(\left(\left(n \cdot 2\right) \cdot \color{blue}{\sqrt{\mathsf{PI}\left(\right)}}\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \]
    13. lift-PI.f64N/A

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(\left(\left(n \cdot 2\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right) \cdot \sqrt{\color{blue}{\mathsf{PI}\left(\right)}}\right)}^{\left(\frac{1 - k}{2}\right)} \]
    14. lower-sqrt.f6499.4

      \[\leadsto \frac{1}{\sqrt{k}} \cdot {\left(\left(\left(n \cdot 2\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right) \cdot \color{blue}{\sqrt{\mathsf{PI}\left(\right)}}\right)}^{\left(\frac{1 - k}{2}\right)} \]
  4. Applied rewrites99.4%

    \[\leadsto \frac{1}{\sqrt{k}} \cdot {\color{blue}{\left(\left(\left(n \cdot 2\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right) \cdot \sqrt{\mathsf{PI}\left(\right)}\right)}}^{\left(\frac{1 - k}{2}\right)} \]
  5. Applied rewrites99.7%

    \[\leadsto \color{blue}{\frac{\sqrt{\mathsf{PI}\left(\right) \cdot \left(n \cdot 2\right)} \cdot 1}{{\left(\mathsf{PI}\left(\right) \cdot \left(n \cdot 2\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}}} \]
  6. Final simplification99.7%

    \[\leadsto \frac{\sqrt{\mathsf{PI}\left(\right) \cdot \left(n \cdot 2\right)}}{{\left(\mathsf{PI}\left(\right) \cdot \left(n \cdot 2\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
  7. Add Preprocessing

Alternative 2: 97.7% accurate, 0.2× speedup?

\[\begin{array}{l} \\ \begin{array}{l} t_0 := {\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}\\ \mathbf{if}\;t\_0 \leq 0:\\ \;\;\;\;0\\ \mathbf{elif}\;t\_0 \leq 5 \cdot 10^{+273}:\\ \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\ \mathbf{else}:\\ \;\;\;\;\frac{{0}^{\left(-0.125 \cdot k\right)}}{\sqrt{k}}\\ \end{array} \end{array} \]
(FPCore (k n)
 :precision binary64
 (let* ((t_0
         (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0)))))
   (if (<= t_0 0.0)
     0.0
     (if (<= t_0 5e+273)
       (* (sqrt (* 2.0 n)) (sqrt (/ (PI) k)))
       (/ (pow 0.0 (* -0.125 k)) (sqrt k))))))
\begin{array}{l}

\\
\begin{array}{l}
t_0 := {\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}\\
\mathbf{if}\;t\_0 \leq 0:\\
\;\;\;\;0\\

\mathbf{elif}\;t\_0 \leq 5 \cdot 10^{+273}:\\
\;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\

\mathbf{else}:\\
\;\;\;\;\frac{{0}^{\left(-0.125 \cdot k\right)}}{\sqrt{k}}\\


\end{array}
\end{array}
Derivation
  1. Split input into 3 regimes
  2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 0.0

    1. Initial program 100.0%

      \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
    2. Add Preprocessing
    3. Applied rewrites1.6%

      \[\leadsto \color{blue}{\frac{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}}} \]
    4. Step-by-step derivation
      1. lift-pow.f64N/A

        \[\leadsto \frac{\color{blue}{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}}{\sqrt{k}} \]
      2. lift-*.f64N/A

        \[\leadsto \frac{{\color{blue}{\left(0 \cdot n\right)}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
      3. mul0-lftN/A

        \[\leadsto \frac{{\color{blue}{0}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
      4. pow-base-0100.0

        \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
    5. Applied rewrites100.0%

      \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
    6. Taylor expanded in k around 0

      \[\leadsto \color{blue}{0} \]
    7. Step-by-step derivation
      1. Applied rewrites100.0%

        \[\leadsto \color{blue}{0} \]

      if 0.0 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 4.99999999999999961e273

      1. Initial program 99.0%

        \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
      2. Add Preprocessing
      3. Taylor expanded in k around 0

        \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
      4. Step-by-step derivation
        1. *-commutativeN/A

          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
        2. lower-*.f64N/A

          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
        3. lower-sqrt.f64N/A

          \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
        4. lower-sqrt.f64N/A

          \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
        5. lower-/.f64N/A

          \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
        6. *-commutativeN/A

          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
        7. lower-*.f64N/A

          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
        8. lower-PI.f6469.6

          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
      5. Applied rewrites69.6%

        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
      6. Step-by-step derivation
        1. Applied rewrites69.7%

          \[\leadsto \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k} \cdot 2}} \]
        2. Step-by-step derivation
          1. Applied rewrites97.1%

            \[\leadsto \sqrt{2 \cdot n} \cdot \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right)}{k}}} \]

          if 4.99999999999999961e273 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

          1. Initial program 100.0%

            \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
          2. Add Preprocessing
          3. Applied rewrites100.0%

            \[\leadsto \color{blue}{\frac{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}}} \]
          4. Taylor expanded in k around inf

            \[\leadsto \frac{{\left(0 \cdot n\right)}^{\color{blue}{\left(\frac{-1}{8} \cdot k\right)}}}{\sqrt{k}} \]
          5. Step-by-step derivation
            1. lower-*.f64100.0

              \[\leadsto \frac{{\left(0 \cdot n\right)}^{\color{blue}{\left(-0.125 \cdot k\right)}}}{\sqrt{k}} \]
          6. Applied rewrites100.0%

            \[\leadsto \frac{{\left(0 \cdot n\right)}^{\color{blue}{\left(-0.125 \cdot k\right)}}}{\sqrt{k}} \]
        3. Recombined 3 regimes into one program.
        4. Final simplification98.5%

          \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{elif}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 5 \cdot 10^{+273}:\\ \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\ \mathbf{else}:\\ \;\;\;\;\frac{{0}^{\left(-0.125 \cdot k\right)}}{\sqrt{k}}\\ \end{array} \]
        5. Add Preprocessing

        Alternative 3: 80.0% accurate, 0.3× speedup?

        \[\begin{array}{l} \\ \begin{array}{l} t_0 := \mathsf{PI}\left(\right) \cdot n\\ t_1 := {\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}\\ \mathbf{if}\;t\_1 \leq 0:\\ \;\;\;\;0\\ \mathbf{elif}\;t\_1 \leq 4 \cdot 10^{+272}:\\ \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\ \mathbf{else}:\\ \;\;\;\;\frac{\left(\sqrt{\sqrt{t\_0}} \cdot \sqrt{\sqrt{k \cdot \left(t\_0 \cdot k\right)}}\right) \cdot \sqrt{2}}{k}\\ \end{array} \end{array} \]
        (FPCore (k n)
         :precision binary64
         (let* ((t_0 (* (PI) n))
                (t_1
                 (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0)))))
           (if (<= t_1 0.0)
             0.0
             (if (<= t_1 4e+272)
               (* (sqrt (* 2.0 n)) (sqrt (/ (PI) k)))
               (/
                (* (* (sqrt (sqrt t_0)) (sqrt (sqrt (* k (* t_0 k))))) (sqrt 2.0))
                k)))))
        \begin{array}{l}
        
        \\
        \begin{array}{l}
        t_0 := \mathsf{PI}\left(\right) \cdot n\\
        t_1 := {\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}\\
        \mathbf{if}\;t\_1 \leq 0:\\
        \;\;\;\;0\\
        
        \mathbf{elif}\;t\_1 \leq 4 \cdot 10^{+272}:\\
        \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\
        
        \mathbf{else}:\\
        \;\;\;\;\frac{\left(\sqrt{\sqrt{t\_0}} \cdot \sqrt{\sqrt{k \cdot \left(t\_0 \cdot k\right)}}\right) \cdot \sqrt{2}}{k}\\
        
        
        \end{array}
        \end{array}
        
        Derivation
        1. Split input into 3 regimes
        2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 0.0

          1. Initial program 100.0%

            \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
          2. Add Preprocessing
          3. Applied rewrites1.6%

            \[\leadsto \color{blue}{\frac{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}}} \]
          4. Step-by-step derivation
            1. lift-pow.f64N/A

              \[\leadsto \frac{\color{blue}{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}}{\sqrt{k}} \]
            2. lift-*.f64N/A

              \[\leadsto \frac{{\color{blue}{\left(0 \cdot n\right)}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
            3. mul0-lftN/A

              \[\leadsto \frac{{\color{blue}{0}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
            4. pow-base-0100.0

              \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
          5. Applied rewrites100.0%

            \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
          6. Taylor expanded in k around 0

            \[\leadsto \color{blue}{0} \]
          7. Step-by-step derivation
            1. Applied rewrites100.0%

              \[\leadsto \color{blue}{0} \]

            if 0.0 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 4.0000000000000003e272

            1. Initial program 99.0%

              \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
            2. Add Preprocessing
            3. Taylor expanded in k around 0

              \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
            4. Step-by-step derivation
              1. *-commutativeN/A

                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
              2. lower-*.f64N/A

                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
              3. lower-sqrt.f64N/A

                \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
              4. lower-sqrt.f64N/A

                \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
              5. lower-/.f64N/A

                \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
              6. *-commutativeN/A

                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
              7. lower-*.f64N/A

                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
              8. lower-PI.f6470.0

                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
            5. Applied rewrites70.0%

              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
            6. Step-by-step derivation
              1. Applied rewrites70.2%

                \[\leadsto \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k} \cdot 2}} \]
              2. Step-by-step derivation
                1. Applied rewrites97.0%

                  \[\leadsto \sqrt{2 \cdot n} \cdot \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right)}{k}}} \]

                if 4.0000000000000003e272 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

                1. Initial program 100.0%

                  \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                2. Add Preprocessing
                3. Taylor expanded in k around 0

                  \[\leadsto \color{blue}{\frac{\frac{-1}{2} \cdot \left(\sqrt{{k}^{3} \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \sqrt{2}\right)\right) + \sqrt{k \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \sqrt{2}}{k}} \]
                4. Step-by-step derivation
                  1. lower-/.f64N/A

                    \[\leadsto \color{blue}{\frac{\frac{-1}{2} \cdot \left(\sqrt{{k}^{3} \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \sqrt{2}\right)\right) + \sqrt{k \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \sqrt{2}}{k}} \]
                5. Applied rewrites72.2%

                  \[\leadsto \color{blue}{\frac{\sqrt{2} \cdot \mathsf{fma}\left(\log \left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right) \cdot \sqrt{\left({k}^{3} \cdot \mathsf{PI}\left(\right)\right) \cdot n}, -0.5, \sqrt{\left(\mathsf{PI}\left(\right) \cdot n\right) \cdot k}\right)}{k}} \]
                6. Taylor expanded in k around 0

                  \[\leadsto \frac{\sqrt{k \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \sqrt{2}}{k} \]
                7. Step-by-step derivation
                  1. Applied rewrites3.7%

                    \[\leadsto \frac{\sqrt{\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot k} \cdot \sqrt{2}}{k} \]
                  2. Step-by-step derivation
                    1. Applied rewrites31.7%

                      \[\leadsto \frac{\left(\sqrt{\sqrt{\mathsf{PI}\left(\right) \cdot n}} \cdot \sqrt{\sqrt{k \cdot \left(\left(\mathsf{PI}\left(\right) \cdot n\right) \cdot k\right)}}\right) \cdot \sqrt{2}}{k} \]
                  3. Recombined 3 regimes into one program.
                  4. Final simplification81.7%

                    \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{elif}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 4 \cdot 10^{+272}:\\ \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\ \mathbf{else}:\\ \;\;\;\;\frac{\left(\sqrt{\sqrt{\mathsf{PI}\left(\right) \cdot n}} \cdot \sqrt{\sqrt{k \cdot \left(\left(\mathsf{PI}\left(\right) \cdot n\right) \cdot k\right)}}\right) \cdot \sqrt{2}}{k}\\ \end{array} \]
                  5. Add Preprocessing

                  Alternative 4: 73.9% accurate, 0.3× speedup?

                  \[\begin{array}{l} \\ \begin{array}{l} t_0 := {\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}\\ \mathbf{if}\;t\_0 \leq 0:\\ \;\;\;\;0\\ \mathbf{elif}\;t\_0 \leq 4 \cdot 10^{+136}:\\ \;\;\;\;\sqrt{\left(2 \cdot n\right) \cdot \frac{\mathsf{PI}\left(\right)}{k}}\\ \mathbf{else}:\\ \;\;\;\;\frac{\sqrt{\left(\left(\mathsf{PI}\left(\right) \cdot n\right) \cdot k\right) \cdot 2}}{k}\\ \end{array} \end{array} \]
                  (FPCore (k n)
                   :precision binary64
                   (let* ((t_0
                           (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0)))))
                     (if (<= t_0 0.0)
                       0.0
                       (if (<= t_0 4e+136)
                         (sqrt (* (* 2.0 n) (/ (PI) k)))
                         (/ (sqrt (* (* (* (PI) n) k) 2.0)) k)))))
                  \begin{array}{l}
                  
                  \\
                  \begin{array}{l}
                  t_0 := {\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}\\
                  \mathbf{if}\;t\_0 \leq 0:\\
                  \;\;\;\;0\\
                  
                  \mathbf{elif}\;t\_0 \leq 4 \cdot 10^{+136}:\\
                  \;\;\;\;\sqrt{\left(2 \cdot n\right) \cdot \frac{\mathsf{PI}\left(\right)}{k}}\\
                  
                  \mathbf{else}:\\
                  \;\;\;\;\frac{\sqrt{\left(\left(\mathsf{PI}\left(\right) \cdot n\right) \cdot k\right) \cdot 2}}{k}\\
                  
                  
                  \end{array}
                  \end{array}
                  
                  Derivation
                  1. Split input into 3 regimes
                  2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 0.0

                    1. Initial program 100.0%

                      \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                    2. Add Preprocessing
                    3. Applied rewrites1.6%

                      \[\leadsto \color{blue}{\frac{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}}} \]
                    4. Step-by-step derivation
                      1. lift-pow.f64N/A

                        \[\leadsto \frac{\color{blue}{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}}{\sqrt{k}} \]
                      2. lift-*.f64N/A

                        \[\leadsto \frac{{\color{blue}{\left(0 \cdot n\right)}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                      3. mul0-lftN/A

                        \[\leadsto \frac{{\color{blue}{0}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                      4. pow-base-0100.0

                        \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                    5. Applied rewrites100.0%

                      \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                    6. Taylor expanded in k around 0

                      \[\leadsto \color{blue}{0} \]
                    7. Step-by-step derivation
                      1. Applied rewrites100.0%

                        \[\leadsto \color{blue}{0} \]

                      if 0.0 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 4.00000000000000023e136

                      1. Initial program 98.9%

                        \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                      2. Add Preprocessing
                      3. Taylor expanded in k around 0

                        \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                      4. Step-by-step derivation
                        1. *-commutativeN/A

                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                        2. lower-*.f64N/A

                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                        3. lower-sqrt.f64N/A

                          \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                        4. lower-sqrt.f64N/A

                          \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                        5. lower-/.f64N/A

                          \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                        6. *-commutativeN/A

                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                        7. lower-*.f64N/A

                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                        8. lower-PI.f6495.8

                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                      5. Applied rewrites95.8%

                        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                      6. Step-by-step derivation
                        1. Applied rewrites96.0%

                          \[\leadsto \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k} \cdot 2}} \]
                        2. Step-by-step derivation
                          1. Applied rewrites96.1%

                            \[\leadsto \sqrt{\left(2 \cdot n\right) \cdot \frac{\mathsf{PI}\left(\right)}{k}} \]

                          if 4.00000000000000023e136 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

                          1. Initial program 99.6%

                            \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                          2. Add Preprocessing
                          3. Taylor expanded in k around 0

                            \[\leadsto \color{blue}{\frac{\frac{-1}{2} \cdot \left(\sqrt{{k}^{3} \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \sqrt{2}\right)\right) + \sqrt{k \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \sqrt{2}}{k}} \]
                          4. Step-by-step derivation
                            1. lower-/.f64N/A

                              \[\leadsto \color{blue}{\frac{\frac{-1}{2} \cdot \left(\sqrt{{k}^{3} \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \sqrt{2}\right)\right) + \sqrt{k \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \sqrt{2}}{k}} \]
                          5. Applied rewrites82.9%

                            \[\leadsto \color{blue}{\frac{\sqrt{2} \cdot \mathsf{fma}\left(\log \left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right) \cdot \sqrt{\left({k}^{3} \cdot \mathsf{PI}\left(\right)\right) \cdot n}, -0.5, \sqrt{\left(\mathsf{PI}\left(\right) \cdot n\right) \cdot k}\right)}{k}} \]
                          6. Taylor expanded in k around 0

                            \[\leadsto \frac{\sqrt{k \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot \sqrt{2}}{k} \]
                          7. Step-by-step derivation
                            1. Applied rewrites41.4%

                              \[\leadsto \frac{\sqrt{\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot k} \cdot \sqrt{2}}{k} \]
                            2. Step-by-step derivation
                              1. Applied rewrites41.5%

                                \[\leadsto \frac{\sqrt{\left(\left(\mathsf{PI}\left(\right) \cdot n\right) \cdot k\right) \cdot 2}}{k} \]
                            3. Recombined 3 regimes into one program.
                            4. Final simplification74.9%

                              \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{elif}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 4 \cdot 10^{+136}:\\ \;\;\;\;\sqrt{\left(2 \cdot n\right) \cdot \frac{\mathsf{PI}\left(\right)}{k}}\\ \mathbf{else}:\\ \;\;\;\;\frac{\sqrt{\left(\left(\mathsf{PI}\left(\right) \cdot n\right) \cdot k\right) \cdot 2}}{k}\\ \end{array} \]
                            5. Add Preprocessing

                            Alternative 5: 73.8% accurate, 0.5× speedup?

                            \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{else}:\\ \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\ \end{array} \end{array} \]
                            (FPCore (k n)
                             :precision binary64
                             (if (<=
                                  (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0)))
                                  0.0)
                               0.0
                               (* (sqrt (* 2.0 n)) (sqrt (/ (PI) k)))))
                            \begin{array}{l}
                            
                            \\
                            \begin{array}{l}
                            \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\
                            \;\;\;\;0\\
                            
                            \mathbf{else}:\\
                            \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\
                            
                            
                            \end{array}
                            \end{array}
                            
                            Derivation
                            1. Split input into 2 regimes
                            2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 0.0

                              1. Initial program 100.0%

                                \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                              2. Add Preprocessing
                              3. Applied rewrites1.6%

                                \[\leadsto \color{blue}{\frac{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}}} \]
                              4. Step-by-step derivation
                                1. lift-pow.f64N/A

                                  \[\leadsto \frac{\color{blue}{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}}{\sqrt{k}} \]
                                2. lift-*.f64N/A

                                  \[\leadsto \frac{{\color{blue}{\left(0 \cdot n\right)}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                                3. mul0-lftN/A

                                  \[\leadsto \frac{{\color{blue}{0}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                                4. pow-base-0100.0

                                  \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                              5. Applied rewrites100.0%

                                \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                              6. Taylor expanded in k around 0

                                \[\leadsto \color{blue}{0} \]
                              7. Step-by-step derivation
                                1. Applied rewrites100.0%

                                  \[\leadsto \color{blue}{0} \]

                                if 0.0 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

                                1. Initial program 99.3%

                                  \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                2. Add Preprocessing
                                3. Taylor expanded in k around 0

                                  \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                4. Step-by-step derivation
                                  1. *-commutativeN/A

                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                  2. lower-*.f64N/A

                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                  3. lower-sqrt.f64N/A

                                    \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                  4. lower-sqrt.f64N/A

                                    \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                  5. lower-/.f64N/A

                                    \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                  6. *-commutativeN/A

                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                  7. lower-*.f64N/A

                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                  8. lower-PI.f6447.8

                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                5. Applied rewrites47.8%

                                  \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                6. Step-by-step derivation
                                  1. Applied rewrites47.9%

                                    \[\leadsto \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k} \cdot 2}} \]
                                  2. Step-by-step derivation
                                    1. Applied rewrites66.4%

                                      \[\leadsto \sqrt{2 \cdot n} \cdot \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right)}{k}}} \]
                                  3. Recombined 2 regimes into one program.
                                  4. Final simplification74.8%

                                    \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{else}:\\ \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\ \end{array} \]
                                  5. Add Preprocessing

                                  Alternative 6: 73.8% accurate, 0.5× speedup?

                                  \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{else}:\\ \;\;\;\;\sqrt{n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot 2}{k}}\\ \end{array} \end{array} \]
                                  (FPCore (k n)
                                   :precision binary64
                                   (if (<=
                                        (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0)))
                                        0.0)
                                     0.0
                                     (* (sqrt n) (sqrt (/ (* (PI) 2.0) k)))))
                                  \begin{array}{l}
                                  
                                  \\
                                  \begin{array}{l}
                                  \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\
                                  \;\;\;\;0\\
                                  
                                  \mathbf{else}:\\
                                  \;\;\;\;\sqrt{n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot 2}{k}}\\
                                  
                                  
                                  \end{array}
                                  \end{array}
                                  
                                  Derivation
                                  1. Split input into 2 regimes
                                  2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 0.0

                                    1. Initial program 100.0%

                                      \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                    2. Add Preprocessing
                                    3. Applied rewrites1.6%

                                      \[\leadsto \color{blue}{\frac{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}}} \]
                                    4. Step-by-step derivation
                                      1. lift-pow.f64N/A

                                        \[\leadsto \frac{\color{blue}{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}}{\sqrt{k}} \]
                                      2. lift-*.f64N/A

                                        \[\leadsto \frac{{\color{blue}{\left(0 \cdot n\right)}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                                      3. mul0-lftN/A

                                        \[\leadsto \frac{{\color{blue}{0}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                                      4. pow-base-0100.0

                                        \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                                    5. Applied rewrites100.0%

                                      \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                                    6. Taylor expanded in k around 0

                                      \[\leadsto \color{blue}{0} \]
                                    7. Step-by-step derivation
                                      1. Applied rewrites100.0%

                                        \[\leadsto \color{blue}{0} \]

                                      if 0.0 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

                                      1. Initial program 99.3%

                                        \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                      2. Add Preprocessing
                                      3. Taylor expanded in k around 0

                                        \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                      4. Step-by-step derivation
                                        1. *-commutativeN/A

                                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                        2. lower-*.f64N/A

                                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                        3. lower-sqrt.f64N/A

                                          \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                        4. lower-sqrt.f64N/A

                                          \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                        5. lower-/.f64N/A

                                          \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                        6. *-commutativeN/A

                                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                        7. lower-*.f64N/A

                                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                        8. lower-PI.f6447.8

                                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                      5. Applied rewrites47.8%

                                        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                      6. Step-by-step derivation
                                        1. Applied rewrites47.9%

                                          \[\leadsto \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k} \cdot 2}} \]
                                        2. Step-by-step derivation
                                          1. Applied rewrites66.4%

                                            \[\leadsto \sqrt{n} \cdot \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right) \cdot 2}{k}}} \]
                                        3. Recombined 2 regimes into one program.
                                        4. Final simplification74.8%

                                          \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{else}:\\ \;\;\;\;\sqrt{n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot 2}{k}}\\ \end{array} \]
                                        5. Add Preprocessing

                                        Alternative 7: 62.8% accurate, 0.5× speedup?

                                        \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{else}:\\ \;\;\;\;\sqrt{\left(2 \cdot n\right) \cdot \frac{\mathsf{PI}\left(\right)}{k}}\\ \end{array} \end{array} \]
                                        (FPCore (k n)
                                         :precision binary64
                                         (if (<=
                                              (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0)))
                                              0.0)
                                           0.0
                                           (sqrt (* (* 2.0 n) (/ (PI) k)))))
                                        \begin{array}{l}
                                        
                                        \\
                                        \begin{array}{l}
                                        \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\
                                        \;\;\;\;0\\
                                        
                                        \mathbf{else}:\\
                                        \;\;\;\;\sqrt{\left(2 \cdot n\right) \cdot \frac{\mathsf{PI}\left(\right)}{k}}\\
                                        
                                        
                                        \end{array}
                                        \end{array}
                                        
                                        Derivation
                                        1. Split input into 2 regimes
                                        2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 0.0

                                          1. Initial program 100.0%

                                            \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                          2. Add Preprocessing
                                          3. Applied rewrites1.6%

                                            \[\leadsto \color{blue}{\frac{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}}} \]
                                          4. Step-by-step derivation
                                            1. lift-pow.f64N/A

                                              \[\leadsto \frac{\color{blue}{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}}{\sqrt{k}} \]
                                            2. lift-*.f64N/A

                                              \[\leadsto \frac{{\color{blue}{\left(0 \cdot n\right)}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                                            3. mul0-lftN/A

                                              \[\leadsto \frac{{\color{blue}{0}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                                            4. pow-base-0100.0

                                              \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                                          5. Applied rewrites100.0%

                                            \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                                          6. Taylor expanded in k around 0

                                            \[\leadsto \color{blue}{0} \]
                                          7. Step-by-step derivation
                                            1. Applied rewrites100.0%

                                              \[\leadsto \color{blue}{0} \]

                                            if 0.0 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

                                            1. Initial program 99.3%

                                              \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                            2. Add Preprocessing
                                            3. Taylor expanded in k around 0

                                              \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                            4. Step-by-step derivation
                                              1. *-commutativeN/A

                                                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                              2. lower-*.f64N/A

                                                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                              3. lower-sqrt.f64N/A

                                                \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                              4. lower-sqrt.f64N/A

                                                \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                              5. lower-/.f64N/A

                                                \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                              6. *-commutativeN/A

                                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                              7. lower-*.f64N/A

                                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                              8. lower-PI.f6447.8

                                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                            5. Applied rewrites47.8%

                                              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                            6. Step-by-step derivation
                                              1. Applied rewrites47.9%

                                                \[\leadsto \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k} \cdot 2}} \]
                                              2. Step-by-step derivation
                                                1. Applied rewrites47.9%

                                                  \[\leadsto \sqrt{\left(2 \cdot n\right) \cdot \frac{\mathsf{PI}\left(\right)}{k}} \]
                                              3. Recombined 2 regimes into one program.
                                              4. Final simplification60.9%

                                                \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;0\\ \mathbf{else}:\\ \;\;\;\;\sqrt{\left(2 \cdot n\right) \cdot \frac{\mathsf{PI}\left(\right)}{k}}\\ \end{array} \]
                                              5. Add Preprocessing

                                              Alternative 8: 99.4% accurate, 0.6× speedup?

                                              \[\begin{array}{l} \\ \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{{k}^{-1}} \end{array} \]
                                              (FPCore (k n)
                                               :precision binary64
                                               (* (sqrt (pow (* (* n (PI)) 2.0) (- 1.0 k))) (sqrt (pow k -1.0))))
                                              \begin{array}{l}
                                              
                                              \\
                                              \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{{k}^{-1}}
                                              \end{array}
                                              
                                              Derivation
                                              1. Initial program 99.5%

                                                \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                              2. Add Preprocessing
                                              3. Taylor expanded in k around inf

                                                \[\leadsto \color{blue}{\sqrt{\frac{1}{k}} \cdot e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)}} \]
                                              4. Applied rewrites99.5%

                                                \[\leadsto \color{blue}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)} \cdot \sqrt{\frac{1}{k}}} \]
                                              5. Taylor expanded in k around inf

                                                \[\leadsto \color{blue}{\sqrt{\frac{1}{k}} \cdot e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)}} \]
                                              6. Step-by-step derivation
                                                1. *-commutativeN/A

                                                  \[\leadsto \color{blue}{e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)} \cdot \sqrt{\frac{1}{k}}} \]
                                                2. lower-*.f64N/A

                                                  \[\leadsto \color{blue}{e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)} \cdot \sqrt{\frac{1}{k}}} \]
                                                3. *-commutativeN/A

                                                  \[\leadsto e^{\color{blue}{\left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right) \cdot \frac{1}{2}}} \cdot \sqrt{\frac{1}{k}} \]
                                                4. exp-prodN/A

                                                  \[\leadsto \color{blue}{{\left(e^{\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)}\right)}^{\frac{1}{2}}} \cdot \sqrt{\frac{1}{k}} \]
                                                5. unpow1/2N/A

                                                  \[\leadsto \color{blue}{\sqrt{e^{\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                6. lower-sqrt.f64N/A

                                                  \[\leadsto \color{blue}{\sqrt{e^{\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                7. exp-to-powN/A

                                                  \[\leadsto \sqrt{\color{blue}{{\left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)}^{\left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                8. lower-pow.f64N/A

                                                  \[\leadsto \sqrt{\color{blue}{{\left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)}^{\left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                9. *-commutativeN/A

                                                  \[\leadsto \sqrt{{\color{blue}{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}} \]
                                                10. lower-*.f64N/A

                                                  \[\leadsto \sqrt{{\color{blue}{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}} \]
                                                11. lower-*.f64N/A

                                                  \[\leadsto \sqrt{{\left(\color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}} \]
                                                12. lower-PI.f64N/A

                                                  \[\leadsto \sqrt{{\left(\left(n \cdot \color{blue}{\mathsf{PI}\left(\right)}\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}} \]
                                                13. lower--.f64N/A

                                                  \[\leadsto \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\color{blue}{\left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                14. lower-sqrt.f64N/A

                                                  \[\leadsto \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \color{blue}{\sqrt{\frac{1}{k}}} \]
                                                15. lower-/.f6499.5

                                                  \[\leadsto \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{\color{blue}{\frac{1}{k}}} \]
                                              7. Applied rewrites99.5%

                                                \[\leadsto \color{blue}{\sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}}} \]
                                              8. Final simplification99.5%

                                                \[\leadsto \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{{k}^{-1}} \]
                                              9. Add Preprocessing

                                              Alternative 9: 99.4% accurate, 0.6× speedup?

                                              \[\begin{array}{l} \\ {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)} \cdot \sqrt{{k}^{-1}} \end{array} \]
                                              (FPCore (k n)
                                               :precision binary64
                                               (* (pow (* (* 2.0 n) (PI)) (fma -0.5 k 0.5)) (sqrt (pow k -1.0))))
                                              \begin{array}{l}
                                              
                                              \\
                                              {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)} \cdot \sqrt{{k}^{-1}}
                                              \end{array}
                                              
                                              Derivation
                                              1. Initial program 99.5%

                                                \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                              2. Add Preprocessing
                                              3. Taylor expanded in k around inf

                                                \[\leadsto \color{blue}{\sqrt{\frac{1}{k}} \cdot e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)}} \]
                                              4. Applied rewrites99.5%

                                                \[\leadsto \color{blue}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)} \cdot \sqrt{\frac{1}{k}}} \]
                                              5. Final simplification99.5%

                                                \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)} \cdot \sqrt{{k}^{-1}} \]
                                              6. Add Preprocessing

                                              Alternative 10: 99.4% accurate, 1.1× speedup?

                                              \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;k \leq 1.25 \cdot 10^{-21}:\\ \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\ \mathbf{else}:\\ \;\;\;\;\sqrt{\frac{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}}{k}}\\ \end{array} \end{array} \]
                                              (FPCore (k n)
                                               :precision binary64
                                               (if (<= k 1.25e-21)
                                                 (* (sqrt (* 2.0 n)) (sqrt (/ (PI) k)))
                                                 (sqrt (/ (pow (* (* n (PI)) 2.0) (- 1.0 k)) k))))
                                              \begin{array}{l}
                                              
                                              \\
                                              \begin{array}{l}
                                              \mathbf{if}\;k \leq 1.25 \cdot 10^{-21}:\\
                                              \;\;\;\;\sqrt{2 \cdot n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k}}\\
                                              
                                              \mathbf{else}:\\
                                              \;\;\;\;\sqrt{\frac{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}}{k}}\\
                                              
                                              
                                              \end{array}
                                              \end{array}
                                              
                                              Derivation
                                              1. Split input into 2 regimes
                                              2. if k < 1.24999999999999993e-21

                                                1. Initial program 99.2%

                                                  \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                                2. Add Preprocessing
                                                3. Taylor expanded in k around 0

                                                  \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                                4. Step-by-step derivation
                                                  1. *-commutativeN/A

                                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                  2. lower-*.f64N/A

                                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                  3. lower-sqrt.f64N/A

                                                    \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                                  4. lower-sqrt.f64N/A

                                                    \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                  5. lower-/.f64N/A

                                                    \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                  6. *-commutativeN/A

                                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                                  7. lower-*.f64N/A

                                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                                  8. lower-PI.f6470.0

                                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                                5. Applied rewrites70.0%

                                                  \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                                6. Step-by-step derivation
                                                  1. Applied rewrites70.2%

                                                    \[\leadsto \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k} \cdot 2}} \]
                                                  2. Step-by-step derivation
                                                    1. Applied rewrites99.5%

                                                      \[\leadsto \sqrt{2 \cdot n} \cdot \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right)}{k}}} \]

                                                    if 1.24999999999999993e-21 < k

                                                    1. Initial program 99.7%

                                                      \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                                    2. Add Preprocessing
                                                    3. Taylor expanded in k around inf

                                                      \[\leadsto \color{blue}{\sqrt{\frac{1}{k}} \cdot e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)}} \]
                                                    4. Applied rewrites99.7%

                                                      \[\leadsto \color{blue}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)} \cdot \sqrt{\frac{1}{k}}} \]
                                                    5. Taylor expanded in k around inf

                                                      \[\leadsto \color{blue}{\sqrt{\frac{1}{k}} \cdot e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)}} \]
                                                    6. Step-by-step derivation
                                                      1. *-commutativeN/A

                                                        \[\leadsto \color{blue}{e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)} \cdot \sqrt{\frac{1}{k}}} \]
                                                      2. lower-*.f64N/A

                                                        \[\leadsto \color{blue}{e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)} \cdot \sqrt{\frac{1}{k}}} \]
                                                      3. *-commutativeN/A

                                                        \[\leadsto e^{\color{blue}{\left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right) \cdot \frac{1}{2}}} \cdot \sqrt{\frac{1}{k}} \]
                                                      4. exp-prodN/A

                                                        \[\leadsto \color{blue}{{\left(e^{\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)}\right)}^{\frac{1}{2}}} \cdot \sqrt{\frac{1}{k}} \]
                                                      5. unpow1/2N/A

                                                        \[\leadsto \color{blue}{\sqrt{e^{\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                      6. lower-sqrt.f64N/A

                                                        \[\leadsto \color{blue}{\sqrt{e^{\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                      7. exp-to-powN/A

                                                        \[\leadsto \sqrt{\color{blue}{{\left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)}^{\left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                      8. lower-pow.f64N/A

                                                        \[\leadsto \sqrt{\color{blue}{{\left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)}^{\left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                      9. *-commutativeN/A

                                                        \[\leadsto \sqrt{{\color{blue}{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}} \]
                                                      10. lower-*.f64N/A

                                                        \[\leadsto \sqrt{{\color{blue}{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}} \]
                                                      11. lower-*.f64N/A

                                                        \[\leadsto \sqrt{{\left(\color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}} \]
                                                      12. lower-PI.f64N/A

                                                        \[\leadsto \sqrt{{\left(\left(n \cdot \color{blue}{\mathsf{PI}\left(\right)}\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}} \]
                                                      13. lower--.f64N/A

                                                        \[\leadsto \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\color{blue}{\left(1 - k\right)}}} \cdot \sqrt{\frac{1}{k}} \]
                                                      14. lower-sqrt.f64N/A

                                                        \[\leadsto \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \color{blue}{\sqrt{\frac{1}{k}}} \]
                                                      15. lower-/.f6499.7

                                                        \[\leadsto \sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{\color{blue}{\frac{1}{k}}} \]
                                                    7. Applied rewrites99.7%

                                                      \[\leadsto \color{blue}{\sqrt{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}} \cdot \sqrt{\frac{1}{k}}} \]
                                                    8. Taylor expanded in k around inf

                                                      \[\leadsto \sqrt{\frac{e^{\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)}}{k}} \]
                                                    9. Step-by-step derivation
                                                      1. Applied rewrites99.7%

                                                        \[\leadsto \sqrt{\frac{{\left(\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2\right)}^{\left(1 - k\right)}}{k}} \]
                                                    10. Recombined 2 regimes into one program.
                                                    11. Add Preprocessing

                                                    Alternative 11: 26.6% accurate, 152.0× speedup?

                                                    \[\begin{array}{l} \\ 0 \end{array} \]
                                                    (FPCore (k n) :precision binary64 0.0)
                                                    double code(double k, double n) {
                                                    	return 0.0;
                                                    }
                                                    
                                                    module fmin_fmax_functions
                                                        implicit none
                                                        private
                                                        public fmax
                                                        public fmin
                                                    
                                                        interface fmax
                                                            module procedure fmax88
                                                            module procedure fmax44
                                                            module procedure fmax84
                                                            module procedure fmax48
                                                        end interface
                                                        interface fmin
                                                            module procedure fmin88
                                                            module procedure fmin44
                                                            module procedure fmin84
                                                            module procedure fmin48
                                                        end interface
                                                    contains
                                                        real(8) function fmax88(x, y) result (res)
                                                            real(8), intent (in) :: x
                                                            real(8), intent (in) :: y
                                                            res = merge(y, merge(x, max(x, y), y /= y), x /= x)
                                                        end function
                                                        real(4) function fmax44(x, y) result (res)
                                                            real(4), intent (in) :: x
                                                            real(4), intent (in) :: y
                                                            res = merge(y, merge(x, max(x, y), y /= y), x /= x)
                                                        end function
                                                        real(8) function fmax84(x, y) result(res)
                                                            real(8), intent (in) :: x
                                                            real(4), intent (in) :: y
                                                            res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
                                                        end function
                                                        real(8) function fmax48(x, y) result(res)
                                                            real(4), intent (in) :: x
                                                            real(8), intent (in) :: y
                                                            res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
                                                        end function
                                                        real(8) function fmin88(x, y) result (res)
                                                            real(8), intent (in) :: x
                                                            real(8), intent (in) :: y
                                                            res = merge(y, merge(x, min(x, y), y /= y), x /= x)
                                                        end function
                                                        real(4) function fmin44(x, y) result (res)
                                                            real(4), intent (in) :: x
                                                            real(4), intent (in) :: y
                                                            res = merge(y, merge(x, min(x, y), y /= y), x /= x)
                                                        end function
                                                        real(8) function fmin84(x, y) result(res)
                                                            real(8), intent (in) :: x
                                                            real(4), intent (in) :: y
                                                            res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
                                                        end function
                                                        real(8) function fmin48(x, y) result(res)
                                                            real(4), intent (in) :: x
                                                            real(8), intent (in) :: y
                                                            res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
                                                        end function
                                                    end module
                                                    
                                                    real(8) function code(k, n)
                                                    use fmin_fmax_functions
                                                        real(8), intent (in) :: k
                                                        real(8), intent (in) :: n
                                                        code = 0.0d0
                                                    end function
                                                    
                                                    public static double code(double k, double n) {
                                                    	return 0.0;
                                                    }
                                                    
                                                    def code(k, n):
                                                    	return 0.0
                                                    
                                                    function code(k, n)
                                                    	return 0.0
                                                    end
                                                    
                                                    function tmp = code(k, n)
                                                    	tmp = 0.0;
                                                    end
                                                    
                                                    code[k_, n_] := 0.0
                                                    
                                                    \begin{array}{l}
                                                    
                                                    \\
                                                    0
                                                    \end{array}
                                                    
                                                    Derivation
                                                    1. Initial program 99.5%

                                                      \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                                    2. Add Preprocessing
                                                    3. Applied rewrites26.0%

                                                      \[\leadsto \color{blue}{\frac{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}}} \]
                                                    4. Step-by-step derivation
                                                      1. lift-pow.f64N/A

                                                        \[\leadsto \frac{\color{blue}{{\left(0 \cdot n\right)}^{\left(\frac{1 - k}{8}\right)}}}{\sqrt{k}} \]
                                                      2. lift-*.f64N/A

                                                        \[\leadsto \frac{{\color{blue}{\left(0 \cdot n\right)}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                                                      3. mul0-lftN/A

                                                        \[\leadsto \frac{{\color{blue}{0}}^{\left(\frac{1 - k}{8}\right)}}{\sqrt{k}} \]
                                                      4. pow-base-026.7

                                                        \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                                                    5. Applied rewrites26.7%

                                                      \[\leadsto \frac{\color{blue}{0}}{\sqrt{k}} \]
                                                    6. Taylor expanded in k around 0

                                                      \[\leadsto \color{blue}{0} \]
                                                    7. Step-by-step derivation
                                                      1. Applied rewrites26.7%

                                                        \[\leadsto \color{blue}{0} \]
                                                      2. Add Preprocessing

                                                      Reproduce

                                                      ?
                                                      herbie shell --seed 2024363 
                                                      (FPCore (k n)
                                                        :name "Migdal et al, Equation (51)"
                                                        :precision binary64
                                                        (* (/ 1.0 (sqrt k)) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0))))