...one of the most highly
regarded and expertly designed C++ library projects in the
world.
— Herb Sutter and Andrei
Alexandrescu, C++
Coding Standards
Random numbers are required in a number of different problem domains, such as
The Boost Random Number Generator Library provides a framework for random number generators with well-defined properties so that the generators can be used in the demanding numerics and security domains. For a general introduction to random numbers in numerics, see
"Numerical Recipes in C: The art of scientific computing", William H. Press, Saul A. Teukolsky, William A. Vetterling, Brian P. Flannery, 2nd ed., 1992, pp. 274-328
Depending on the requirements of the problem domain, different variations of random number generators are appropriate:
All variations have some properties in common, the concepts (in the STL sense) is called UniformRandomNumberGenerator. This concept will be defined in a subsequent section.
The goals for this library are the following:
A uniform random number generator provides a sequence of random numbers uniformly distributed on a given range. The range can be compile-time fixed or available (only) after run-time construction of the object.
The tight lower bound of some (finite) set S is the (unique) member l in S, so that for all v in S, l <= v holds. Likewise, the tight upper bound of some (finite) set S is the (unique) member u in S, so that for all v in S, v <= u holds.
In the following table, X denotes a number generator class returning objects of type T, and v is a const value of X.
Table 21.1. UniformRandomNumberGenerator requirements
expression |
return type |
pre/post-condition |
---|---|---|
|
|
|
|
|
- |
|
|
tight lower bound on the set of all values returned by |
|
|
if |
The member functions min
,
max
, and operator()
shall have amortized constant time complexity.
Note | |
---|---|
For integer generators (i.e. integer Rationale: The range description with min and max serves two purposes. First, it allows scaling of the values to some canonical range, such as [0..1). Second, it describes the significant bits of the values, which may be relevant for further processing. The range is a closed interval [min,max] for integers, because the underlying type may not be able to represent the half-open interval [min,max+1). It is a half-open interval [min, max) for non-integers, because this is much more practical for borderline cases of continuous distributions. |
Note | |
---|---|
The UniformRandomNumberGenerator
concept does not require
Rationale: |
A non-deterministic uniform random number generator is a UniformRandomNumberGenerator that is based on some stochastic process. Thus, it provides a sequence of truly-random numbers. Examples for such processes are nuclear decay, noise of a Zehner diode, tunneling of quantum particles, rolling a die, drawing from an urn, and tossing a coin. Depending on the environment, inter-arrival times of network packets or keyboard events may be close approximations of stochastic processes.
The class random_device
is a model for a non-deterministic random number generator.
Note | |
---|---|
This type of random-number generator is useful for security applications, where it is important to prevent an outside attacker from guessing the numbers and thus obtaining your encryption or authentication key. Thus, models of this concept should be cautious not to leak any information, to the extent possible by the environment. For example, it might be advisable to explicitly clear any temporary storage as soon as it is no longer needed. |
A pseudo-random number generator is a UniformRandomNumberGenerator
which provides a deterministic sequence of pseudo-random numbers, based
on some algorithm and internal state. Linear
congruential
and inversive
congruential
generators are examples of such pseudo-random
number generators. Often, these generators are very sensitive to
their parameters. In order to prevent wrong implementations from being
used, an external testsuite should check that the generated sequence and
the validation value provided do indeed match.
Donald E. Knuth gives an extensive overview on pseudo-random number generation in his book "The Art of Computer Programming, Vol. 2, 3rd edition, Addison-Wesley, 1997". The descriptions for the specific generators contain additional references.
Note | |
---|---|
Because the state of a pseudo-random number generator is necessarily finite, the sequence of numbers returned by the generator will loop eventually. |
In addition to the UniformRandomNumberGenerator
requirements, a pseudo-random number generator has some additional requirements.
In the following table, X
denotes a pseudo-random number generator class returning objects of type
T
, x
is a value of T
, u
is a value of X
,
and v
is a const value
of X
.
Table 21.2. PseudoRandomNumberGenerator requirements
expression |
return type |
pre/post-condition |
---|---|---|
|
- |
creates a generator in some implementation-defined state. Note: Several generators thusly created may possibly produce dependent or identical sequences of random numbers. |
|
- |
creates a generator with user-provided state; the implementation shall specify the constructor argument(s) |
|
|
sets the current state according to the argument(s); at least functions with the same signature as the non-default constructor(s) shall be provided. |
Note | |
---|---|
The seed member function is similar to the assign member function in STL containers. However, the naming did not seem appropriate. |
Classes which model a pseudo-random number generator shall also model
EqualityComparable,
i.e. implement operator==
.
Two pseudo-random number generators are defined to be equivalent
if they both return an identical sequence of numbers starting from a given
state.
Classes which model a pseudo-random number generator should also model
the Streamable concept, i.e. implement operator<<
and operator>>
. If so, operator<<
writes all current state of the
pseudo-random number generator to the given ostream
so that operator>>
can restore the state at a later time. The state shall be written in a
platform-independent manner, but it is assumed that the locales
used for writing and reading be the same. The pseudo-random number generator
with the restored state and the original at the just-written state shall
be equivalent.
Classes which model a pseudo-random number generator should also model the CopyConstructible and Assignable concepts. However, note that the sequences of the original and the copy are strongly correlated (in fact, they are identical), which may make them unsuitable for some problem domains. Thus, copying pseudo-random number generators is discouraged; they should always be passed by (non-const) reference.
The classes rand48
,
minstd_rand
, and
mt19937
are models
for a pseudo-random number generator.
Note | |
---|---|
This type of random-number generator is useful for numerics, games and
testing. The non-zero arguments constructor(s) and the |
A random distribution produces random numbers distributed according to
some distribution, given uniformly distributed random values as input.
In the following table, X
denotes a random distribution class returning objects of type T
, u
is a value of X
, x
is a (possibly const) value of X
, and e
is an lvalue of an arbitrary type that meets the requirements of a UniformRandomNumberGenerator,
returning values of type U
.
Table 21.3. Random distribution requirements (in addition to CopyConstructible, and Assignable)
expression |
return type |
pre/post-condition |
complexity |
---|---|---|---|
|
|
- |
compile-time |
|
|
subsequent uses of |
constant |
|
|
the sequence of numbers returned by successive invocations with
the same object |
amortized constant number of invocations of |
|
|
writes a textual representation for the parameters and additional
internal data of the distribution |
O(size of state) |
|
|
restores the parameters and additional internal data of the distribution
|
O(size of state) |
Additional requirements: The sequence of numbers produced by repeated invocations
of x(e)
does
not change whether or not os
<< x
is invoked between any of the invocations x(e)
.
If a textual representation is written using os
<< x
and that representation is restored into the same or a different object
y
of the same type using
is >>
y
, repeated invocations of y(e)
produce the same sequence of random numbers
as would repeated invocations of x(e)
.
This library provides several pseudo-random
number generators. The quality of a pseudo
random number generator crucially depends on both the algorithm and
its parameters. This library implements the algorithms as class templates
with template value parameters, hidden in namespace
boost::random
. Any particular choice of parameters
is represented as the appropriately specializing typedef
in namespace boost
.
Pseudo-random number generators should not be constructed (initialized) frequently during program execution, for two reasons. First, initialization requires full initialization of the internal state of the generator. Thus, generators with a lot of internal state (see below) are costly to initialize. Second, initialization always requires some value used as a "seed" for the generated sequence. It is usually difficult to obtain several good seed values. For example, one method to obtain a seed is to determine the current time at the highest resolution available, e.g. microseconds or nanoseconds. When the pseudo-random number generator is initialized again with the then-current time as the seed, it is likely that this is at a near-constant (non-random) distance from the time given as the seed for first initialization. The distance could even be zero if the resolution of the clock is low, thus the generator re-iterates the same sequence of random numbers. For some applications, this is inappropriate.
Note that all pseudo-random
number generators described below are CopyConstructible
and Assignable. Copying
or assigning a generator will copy all its internal state, so the original
and the copy will generate the identical sequence of random numbers. Often,
such behavior is not wanted. In particular, beware of the algorithms from
the standard library such as std::generate
.
They take a functor argument by value, thereby invoking the copy constructor
when called.
The following table gives an overview of some characteristics of the generators. The cycle length is a rough estimate of the quality of the generator; the approximate relative speed is a performance measure, higher numbers mean faster random number generation.
Table 21.4. generators
generator |
length of cycle |
approx. memory requirements |
approx. speed compared to fastest |
comment |
---|---|---|---|---|
231-2 |
|
16% |
- |
|
231-2 |
|
16% |
- |
|
248-1 |
|
64% |
- |
|
approx. 261 |
|
7% |
- |
|
? |
|
12% |
- |
|
? |
|
37% |
- |
|
~288 |
|
100% |
- |
|
231-1 |
|
2% |
good uniform distribution in several dimensions |
|
211213-1 |
|
100% |
good uniform distribution in up to 350 dimensions |
|
219937-1 |
|
93% |
good uniform distribution in up to 623 dimensions |
|
219937-1 |
|
38% |
good uniform distribution in up to 311 dimensions |
|
~232000 |
|
59% |
- |
|
~267000 |
|
59% |
- |
|
~2120000 |
|
61% |
- |
|
~2170000 |
|
62% |
- |
|
~2230000 |
|
59% |
- |
|
~2510000 |
|
61% |
- |
|
~21050000 |
|
59% |
- |
|
~21200000 |
|
61% |
- |
|
~22300000 |
|
59% |
- |
|
~10171 |
|
5% |
- |
|
~10171 |
|
3% |
- |
|
~10171 |
|
5% |
- |
|
~10171 |
|
3% |
- |
|
~10171 |
|
5% |
- |
|
~10171 |
|
3% |
- |
|
~10171 |
|
5% |
- |
|
~10171 |
|
3% |
- |
|
~10171 |
|
5% |
- |
|
~10171 |
|
3% |
- |
As observable from the table, there is generally a quality/performance/memory trade-off to be decided upon when choosing a random-number generator. The multitude of generators provided in this library allows the application programmer to optimize the trade-off with regard to his application domain. Additionally, employing several fundamentally different random number generators for a given application of Monte Carlo simulation will improve the confidence in the results.
If the names of the generators don't ring any bell and you have no idea which
generator to use, it is reasonable to employ mt19937
for a start: It is fast and has acceptable quality.
Note | |
---|---|
These random number generators are not intended for use in applications
where non-deterministic random numbers are required. See |
In addition to the random number generators, this library provides distribution functions which map one distribution (often a uniform distribution provided by some generator) to another.
Usually, there are several possible implementations of any given mapping. Often, there is a choice between using more space, more invocations of the underlying source of random numbers, or more time-consuming arithmetic such as trigonometric functions. This interface description does not mandate any specific implementation. However, implementations which cannot reach certain values of the specified distribution or otherwise do not converge statistically to it are not acceptable.
Table 21.5. Uniform Distributions
distribution |
explanation |
example |
---|---|---|
discrete uniform distribution on a small set of integers (much smaller than the range of the underlying generator) |
drawing from an urn |
|
discrete uniform distribution on a set of integers; the underlying generator may be called several times to gather enough randomness for the output |
drawing from an urn |
|
continuous uniform distribution on the range [0,1); important basis for other distributions |
- |
|
continuous uniform distribution on some range [min, max) of real numbers |
for the range [0, 2pi): randomly dropping a stick and measuring its angle in radians (assuming the angle is uniformly distributed) |
Table 21.6. Bernoulli Distributions
distribution |
explanation |
example |
---|---|---|
Bernoulli experiment: discrete boolean valued distribution with configurable probability |
tossing a coin (p=0.5) |
|
counts outcomes of repeated Bernoulli experiments |
tossing a coin 20 times and counting how many front sides are shown |
|
measures distance between outcomes of repeated Bernoulli experiments |
throwing a die several times and counting the number of tries until a "6" appears for the first time |
|
Counts the number of failures of repeated Bernoulli experiments required to get some constant number of successes. |
flipping a coin and counting the number of heads that show up before we get 3 tails |
Table 21.7. Poisson Distributions
distribution |
explanation |
example |
---|---|---|
poisson distribution |
counting the number of alpha particles emitted by radioactive matter in a fixed period of time |
|
exponential distribution |
measuring the inter-arrival time of alpha particles emitted by radioactive matter |
|
gamma distribution |
- |
|
weibull distribution |
- |
|
extreme value distribution |
- |
Table 21.8. Normal Distributions
distribution |
explanation |
example |
---|---|---|
counts outcomes of (infinitely) repeated Bernoulli experiments |
tossing a coin 10000 times and counting how many front sides are shown |
|
lognormal distribution (sometimes used in simulations) |
measuring the job completion time of an assembly line worker |
|
chi-squared distribution |
- |
|
Cauchy distribution |
- |
|
Fisher F distribution |
- |
|
Student t distribution |
- |
Table 21.9. Sampling Distributions
distribution |
explanation |
example |
---|---|---|
discrete distribution with specific probabilities |
rolling an unfair die |
|
- |
- |
|
- |
- |
Table 21.10. Miscellaneous Distributions
distribution |
explanation |
example |
---|---|---|
triangle distribution |
- |
|
uniform distribution on a unit sphere of arbitrary dimension |
choosing a random point on Earth (assumed to be a sphere) where to spend the next vacations |
namespace boost { namespace random { template<typename MLCG1, typename MLCG2> class additive_combine_engine; typedef additive_combine_engine< linear_congruential_engine< uint32_t, 40014, 0, 2147483563 >, linear_congruential_engine< uint32_t, 40692, 0, 2147483399 >> ecuyer1988; } }
namespace boost { namespace random { template<typename RealType = double> class bernoulli_distribution; } }
namespace boost { namespace random { template<typename IntType = int, typename RealType = double> class binomial_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class cauchy_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class chi_squared_distribution; } }
namespace boost { namespace random { template<typename UniformRandomNumberGenerator, std::size_t p, std::size_t r> class discard_block_engine; } }
namespace boost { namespace random { template<typename IntType = int, typename WeightType = double> class discrete_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class exponential_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class extreme_value_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class fisher_f_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class gamma_distribution; } }
namespace boost { namespace random { template<typename IntType = int, typename RealType = double> class geometric_distribution; } }
namespace boost { namespace random { template<typename Engine, std::size_t w, typename UIntType> class independent_bits_engine; } }
namespace boost { namespace random { template<typename IntType, IntType a, IntType b, IntType p> class inversive_congruential_engine; typedef inversive_congruential_engine< uint32_t, 9102, 2147483647-36884165, 2147483647 > hellekalek1995; } }
namespace boost { namespace random { template<typename UIntType, int w, unsigned int p, unsigned int q> class lagged_fibonacci_engine; template<typename RealType, int w, unsigned int p, unsigned int q> class lagged_fibonacci_01_engine; typedef lagged_fibonacci_01_engine< double, 48, 607, 273 > lagged_fibonacci607; typedef lagged_fibonacci_01_engine< double, 48, 1279, 418 > lagged_fibonacci1279; typedef lagged_fibonacci_01_engine< double, 48, 2281, 1252 > lagged_fibonacci2281; typedef lagged_fibonacci_01_engine< double, 48, 3217, 576 > lagged_fibonacci3217; typedef lagged_fibonacci_01_engine< double, 48, 4423, 2098 > lagged_fibonacci4423; typedef lagged_fibonacci_01_engine< double, 48, 9689, 5502 > lagged_fibonacci9689; typedef lagged_fibonacci_01_engine< double, 48, 19937, 9842 > lagged_fibonacci19937; typedef lagged_fibonacci_01_engine< double, 48, 23209, 13470 > lagged_fibonacci23209; typedef lagged_fibonacci_01_engine< double, 48, 44497, 21034 > lagged_fibonacci44497; } }
namespace boost { namespace random { template<typename IntType, IntType a, IntType c, IntType m> class linear_congruential_engine; class rand48; typedef linear_congruential_engine< uint32_t, 16807, 0, 2147483647 > minstd_rand0; typedef linear_congruential_engine< uint32_t, 48271, 0, 2147483647 > minstd_rand; } }
namespace boost { namespace random { template<typename UIntType, int w, int k, int q, int s> class linear_feedback_shift_engine; } }
namespace boost { namespace random { template<typename RealType = double> class lognormal_distribution; } }
namespace boost { namespace random { template<typename UIntType, std::size_t w, std::size_t n, std::size_t m, std::size_t r, UIntType a, std::size_t u, UIntType d, std::size_t s, UIntType b, std::size_t t, UIntType c, std::size_t l, UIntType f> class mersenne_twister_engine; typedef mersenne_twister_engine< uint32_t, 32, 351, 175, 19, 0xccab8ee7, 11, 0xffffffff, 7, 0x31b6ab00, 15, 0xffe50000, 17, 1812433253 > mt11213b; typedef mersenne_twister_engine< uint32_t, 32, 624, 397, 31, 0x9908b0df, 11, 0xffffffff, 7, 0x9d2c5680, 15, 0xefc60000, 18, 1812433253 > mt19937; typedef mersenne_twister_engine< uint64_t, 64, 312, 156, 31, 0xb5026f5aa96619e9ull, 29, 0x5555555555555555ull, 17, 0x71d67fffeda60000ull, 37, 0xfff7eee000000000ull, 43, 6364136223846793005ull > mt19937_64; } }
namespace boost { namespace random { template<typename IntType = int, typename RealType = double> class negative_binomial_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class normal_distribution; } }
namespace boost { namespace random { template<typename RealType = double, typename WeightType = double> class piecewise_constant_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class piecewise_linear_distribution; } }
namespace boost { namespace random { template<typename IntType = int, typename RealType = double> class poisson_distribution; } }
namespace boost { namespace random { class random_device; } }
namespace boost { namespace random { template<typename URNG, typename IntType = long> class random_number_generator; } }
namespace boost { namespace random { typedef subtract_with_carry_engine< uint32_t, 24, 10, 24 > ranlux_base; typedef subtract_with_carry_01_engine< float, 24, 10, 24 > ranlux_base_01; typedef subtract_with_carry_01_engine< double, 48, 10, 24 > ranlux64_base_01; typedef discard_block_engine< ranlux_base, 223, 24 > ranlux3; typedef discard_block_engine< ranlux_base, 389, 24 > ranlux4; typedef discard_block_engine< ranlux_base_01, 223, 24 > ranlux3_01; typedef discard_block_engine< ranlux_base_01, 389, 24 > ranlux4_01; typedef discard_block_engine< ranlux64_base_01, 223, 24 > ranlux64_3_01; typedef discard_block_engine< ranlux64_base_01, 389, 24 > ranlux64_4_01; typedef subtract_with_carry_engine< uint64_t, 48, 10, 24 > ranlux64_base; typedef discard_block_engine< ranlux64_base, 223, 24 > ranlux64_3; typedef discard_block_engine< ranlux64_base, 389, 24 > ranlux64_4; typedef subtract_with_carry_engine< uint32_t, 24, 10, 24 > ranlux24_base; typedef subtract_with_carry_engine< uint64_t, 48, 5, 12 > ranlux48_base; typedef discard_block_engine< ranlux24_base, 223, 23 > ranlux24; typedef discard_block_engine< ranlux48_base, 389, 11 > ranlux48; } }
namespace boost { namespace random { class seed_seq; } }
namespace boost { namespace random { template<typename UniformRandomNumberGenerator, std::size_t k> class shuffle_order_engine; typedef shuffle_order_engine< linear_congruential_engine< uint32_t, 1366, 150889, 714025 >, 97 > kreutzer1986; typedef shuffle_order_engine< minstd_rand0, 256 > knuth_b; } }
namespace boost { namespace random { template<typename RealType = double> class student_t_distribution; } }
namespace boost { namespace random { template<typename IntType, std::size_t w, std::size_t s, std::size_t r> class subtract_with_carry_engine; template<typename RealType, std::size_t w, std::size_t s, std::size_t r> class subtract_with_carry_01_engine; } }
namespace boost { namespace random { typedef xor_combine_engine< xor_combine_engine< linear_feedback_shift_engine< uint32_t, 32, 31, 13, 12 >, 0, linear_feedback_shift_engine< uint32_t, 32, 29, 2, 4 >, 0 >, 0, linear_feedback_shift_engine< uint32_t, 32, 28, 3, 17 >, 0 > taus88; } }
namespace boost { namespace random { template<typename RealType = double> class triangle_distribution; } }
namespace boost { namespace random { template<typename RealType = double> class uniform_01; } }
namespace boost { namespace random { template<typename IntType = int> class uniform_int_distribution; } }
namespace boost { namespace random { template<typename RealType = double, typename Cont = std::vector<RealType> > class uniform_on_sphere; } }
namespace boost { namespace random { template<typename RealType = double> class uniform_real_distribution; } }
namespace boost { namespace random { template<typename IntType = int> class uniform_smallint; } }
namespace boost { template<typename Engine, typename Distribution> class variate_generator; }
namespace boost { namespace random { template<typename RealType = double> class weibull_distribution; } }
namespace boost { namespace random { template<typename URNG1, int s1, typename URNG2, int s2> class xor_combine_engine; } }