Data.SBV.Examples.Misc.Floating

Several examples involving IEEE-754 floating point numbers, i.e., single precision Float (SFloat) and double precision Double (SDouble) types.

Note that arithmetic with floating point is full of surprises; due to precision issues associativity of arithmetic operations typically do not hold. Also, the presence of NaN is always something to look out for.

FP addition is not associative

assocPlus :: SFloat -> SFloat -> SFloat -> SBool

Prove that floating point addition is not associative. For illustration purposes, we will require one of the inputs to be a NaN. We have:

>>> prove $ assocPlus (0/0)
Falsifiable. Counter-example:
  s0 = 0.0 :: Float
  s1 = 0.0 :: Float

Indeed:

>>> let i = 0/0 :: Float
>>> i + (0.0 + 0.0)
NaN
>>> ((i + 0.0) + 0.0)
NaN

But keep in mind that NaN does not equal itself in the floating point world! We have:

>>> let nan = 0/0 :: Float in nan == nan
False

assocPlusRegular :: IO ThmResult

Prove that addition is not associative, even if we ignore NaN/Infinity values. To do this, we use the predicate fpIsPoint, which is true of a floating point number (SFloat or SDouble) if it is neither NaN nor Infinity. (That is, it's a representable point in the real-number line.)

We have:

>>> assocPlusRegular
Falsifiable. Counter-example:
  x = -1.5991211e-2 :: Float
  y = 131071.99 :: Float
  z = -131069.99 :: Float

Indeed, we have:

>>> ((-1.5991211e-2) + (131071.99 + (-131069.99))) :: Float
1.9840088
>>> ((-1.5991211e-2) + 131071.99) + (-131069.99) :: Float
1.984375

Note the significant difference between two additions!

FP addition by non-zero can result in no change

nonZeroAddition :: IO ThmResult

Demonstrate that a+b = a does not necessarily mean b is 0 in the floating point world, even when we disallow the obvious solution when a and b are Infinity. We have:

>>> nonZeroAddition
Falsifiable. Counter-example:
  a = 5.1705105e-26 :: Float
  b = -3.8518597e-34 :: Float

Indeed, we have:

>>> (5.1705105e-26 + (-3.8518597e-34)) == (5.1705105e-26 :: Float)
True

But:

>>> -3.8518597e-34 == (0::Float)
False

FP multiplicative inverses may not exist

multInverse :: IO ThmResult

This example illustrates that a * (1/a) does not necessarily equal 1. Again, we protect against division by 0 and NaN/Infinity.

We have:

>>> multInverse
Falsifiable. Counter-example:
  a = 1.1058928764217435e308 :: Double

Indeed, we have:

>>> let a = 1.1058928764217435e308 :: Double
>>> a * (1/a)
0.9999999999999998

Effect of rounding modes

roundingAdd :: IO SatResult

One interesting aspect of floating-point is that the chosen rounding-mode can effect the results of a computation if the exact result cannot be precisely represented. SBV exports the functions fpAdd, fpSub, fpMul, fpDiv, fpFMA and fpSqrt which allows users to specify the IEEE supported RoundingMode for the operation. (Also see the class RoundingFloat.) This example illustrates how SBV can be used to find rounding-modes where, for instance, addition can produce different results. We have:

>>> roundingAdd
Satisfiable. Model:
  rm = RoundNearestTiesToAway :: RoundingMode
  x = 2.0644195e19 :: Float
  y = -2.1974389e18 :: Float

Unfortunately we can't directly validate this result at the Haskell level, as Haskell only supports RoundNearestTiesToEven. We have:

>>> (2.0644195e19 + (-2.1974389e18)) :: Float
1.8446757e19

While we cannot directly see the result when the mode is RoundNearestTiesToAway in Haskell, we can use SBV to provide us with that result thusly:

>>> sat $ \z -> z .== fpAdd sRoundNearestTiesToAway 2.0644195e19 (-2.1974389e18 :: SFloat)
Satisfiable. Model:
  s0 = 1.8446755e19 :: Float

We can see why these two resuls are indeed different. To see why, one would have to convert the individual numbers to Float's, which would induce rounding-errors, add them up, and round-back; a tedious operation, but one that might prove illimunating for the interested reader. We'll merely note that floating point representation and semantics is indeed a thorny subject, and point to https://ece.uwaterloo.ca/~dwharder/NumericalAnalysis/02Numerics/Double/paper.pdf as an excellent guide.