2frac (problem 3.3.1)

Average Accuracy: 77.2% → 99.9%

Time: 6.1s

Precision: binary64

Cost: 448

Math

\[\frac{1}{x + 1} - \frac{1}{x} \]

↓

\[\frac{\frac{1}{-1 - x}}{x} \]

FPCore

(FPCore (x) :precision binary64 (- (/ 1.0 (+ x 1.0)) (/ 1.0 x)))

↓

(FPCore (x) :precision binary64 (/ (/ 1.0 (- -1.0 x)) x))

C

double code(double x) {
	return (1.0 / (x + 1.0)) - (1.0 / x);
}

↓

double code(double x) {
	return (1.0 / (-1.0 - x)) / x;
}

Fortran

real(8) function code(x)
    real(8), intent (in) :: x
    code = (1.0d0 / (x + 1.0d0)) - (1.0d0 / x)
end function

↓

real(8) function code(x)
    real(8), intent (in) :: x
    code = (1.0d0 / ((-1.0d0) - x)) / x
end function

Java

public static double code(double x) {
	return (1.0 / (x + 1.0)) - (1.0 / x);
}

↓

public static double code(double x) {
	return (1.0 / (-1.0 - x)) / x;
}

Python

def code(x):
	return (1.0 / (x + 1.0)) - (1.0 / x)

↓

def code(x):
	return (1.0 / (-1.0 - x)) / x

Julia

function code(x)
	return Float64(Float64(1.0 / Float64(x + 1.0)) - Float64(1.0 / x))
end

↓

function code(x)
	return Float64(Float64(1.0 / Float64(-1.0 - x)) / x)
end

MATLAB

function tmp = code(x)
	tmp = (1.0 / (x + 1.0)) - (1.0 / x);
end

↓

function tmp = code(x)
	tmp = (1.0 / (-1.0 - x)) / x;
end

Wolfram

code[x_] := N[(N[(1.0 / N[(x + 1.0), $MachinePrecision]), $MachinePrecision] - N[(1.0 / x), $MachinePrecision]), $MachinePrecision]

↓

code[x_] := N[(N[(1.0 / N[(-1.0 - x), $MachinePrecision]), $MachinePrecision] / x), $MachinePrecision]

Derivation

1. Initial program 77.2%

   \[\frac{1}{x + 1} - \frac{1}{x} \]

2. Applied egg-rr 78.3%

   \[\leadsto \color{blue}{\frac{\frac{x + \left(-1 - x\right)}{1 + x}}{x}} \]

3. Applied egg-rr 99.9%

   \[\leadsto \frac{\color{blue}{\left(1 + \left(x - x\right)\right) \cdot \frac{1}{-1 - x}}}{x} \]

4. Simplified 99.9%

   \[\leadsto \frac{\color{blue}{\frac{1}{-1 - x}}}{x} \]
   Proof

   [Start] 99.9	\[ \frac{\left(1 + \left(x - x\right)\right) \cdot \frac{1}{-1 - x}}{x} \]
   +-commutative [=>] 99.9	\[ \frac{\color{blue}{\left(\left(x - x\right) + 1\right)} \cdot \frac{1}{-1 - x}}{x} \]
   +-inverses [=>] 99.9	\[ \frac{\left(\color{blue}{0} + 1\right) \cdot \frac{1}{-1 - x}}{x} \]
   metadata-eval [=>] 99.9	\[ \frac{\color{blue}{1} \cdot \frac{1}{-1 - x}}{x} \]
   *-lft-identity [=>] 99.9	\[ \frac{\color{blue}{\frac{1}{-1 - x}}}{x} \]

5. Final simplification 99.9%

   \[\leadsto \frac{\frac{1}{-1 - x}}{x} \]

Alternatives

Alternative 1
Accuracy	98.4%
Cost	712

\[\begin{array}{l} \mathbf{if}\;x \leq -1:\\ \;\;\;\;\frac{\frac{-1}{x}}{x}\\ \mathbf{elif}\;x \leq 1:\\ \;\;\;\;\left(1 - x\right) + \frac{-1}{x}\\ \mathbf{else}:\\ \;\;\;\;\frac{-1}{x \cdot x}\\ \end{array} \]

Alternative 2
Accuracy	98.0%
Cost	585

\[\begin{array}{l} \mathbf{if}\;x \leq -1 \lor \neg \left(x \leq 0.77\right):\\ \;\;\;\;\frac{-1}{x \cdot x}\\ \mathbf{else}:\\ \;\;\;\;1 + \frac{-1}{x}\\ \end{array} \]

Alternative 3
Accuracy	97.9%
Cost	585

\[\begin{array}{l} \mathbf{if}\;x \leq -1 \lor \neg \left(x \leq 0.77\right):\\ \;\;\;\;\frac{-1}{x \cdot x}\\ \mathbf{else}:\\ \;\;\;\;\frac{-1 + x}{x}\\ \end{array} \]

Alternative 4
Accuracy	98.2%
Cost	584

\[\begin{array}{l} \mathbf{if}\;x \leq -1:\\ \;\;\;\;\frac{\frac{-1}{x}}{x}\\ \mathbf{elif}\;x \leq 0.77:\\ \;\;\;\;\frac{-1 + x}{x}\\ \mathbf{else}:\\ \;\;\;\;\frac{-1}{x \cdot x}\\ \end{array} \]

Alternative 5
Accuracy	99.5%
Cost	448

\[\frac{-1}{x + x \cdot x} \]

Alternative 6
Accuracy	51.4%
Cost	192

\[\frac{-1}{x} \]

Alternative 7
Accuracy	3.2%
Cost	128

\[-x \]

Alternative 8
Accuracy	3.0%
Cost	64

\[1 \]

Reproduce

herbie shell --seed 2023141 
(FPCore (x)
  :name "2frac (problem 3.3.1)"
  :precision binary64
  (- (/ 1.0 (+ x 1.0)) (/ 1.0 x)))