tfp.substrates.numpy.distributions.MultivariateNormalDiagPlusLowRank

The multivariate normal distribution on R^k.

Inherits From: MultivariateNormalLinearOperator

The Multivariate Normal distribution is defined over R^k and parameterized by a (batch of) length-k loc vector (aka 'mu') and a (batch of) k x k scale matrix; covariance = scale @ scale.T where @ denotes matrix-multiplication.

Note that this is not the usual construction of a multivariate normal with a diagonal plus low-rank covariance matrix. Instead the scale matrix is diagonal plus low-rank. This means

C = S S.T
  = (D + U E U.T)(D + U E U.T)
  = D**2 + D U E U.T + U E U.T D + (U E U.T)(U E U.T)

where

  • C is the NxN covariance matrix,
  • S is the NxN scale matrix,
  • D is the NxN diagonal part of the scale matrix,
  • U is the NxK low-rank update part of the scale matrix,
  • E is the KxK diagonal inside the low-rank update, and
  • M.T is matrix transpose.

A covariance matrix in this form is not typically itself easily expressible as a diagonal plus low-rank.

Mathematical Details

The probability density function (pdf) is,

pdf(x; loc, scale) = exp(-0.5 ||y||**2) / Z,
y = inv(scale) @ (x - loc),
Z = (2 pi)**(0.5 k) |det(scale)|,

where:

  • loc is a vector in R^k,
  • scale is a linear operator in R^{k x k}, cov = scale @ scale.T,
  • Z denotes the normalization constant, and,
  • ||y||**2 denotes the squared Euclidean norm of y.

A (non-batch) scale matrix is:

scale = diag(scale_diag) +
      scale_perturb_factor @ diag(scale_perturb_diag) @ scale_perturb_factor.T

where:

  • scale_diag.shape = [k],
  • scale_perturb_factor.shape = [k, r], typically k >> r, and,
  • scale_perturb_diag.shape = [r].

Additional leading dimensions (if any) will index batches.

If scale_diag is None, then it defaults to the Identity matrix.

The MultivariateNormal distribution is a member of the location-scale family, i.e., it can be constructed as,

X ~ MultivariateNormal(loc=0, scale=1)   # Identity scale, zero shift.
Y = scale @ X + loc

Examples

tfd = tfp.distributions

# Initialize a single 3-variate Gaussian with covariance `cov = S @ S.T`,
# `S = diag(d) + U @ diag(m) @ U.T`. The perturbation, `U @ diag(m) @ U.T`, is
# a rank-2 update.
mu = [-0.5., 0, 0.5]   # shape: [3]
d = [1.5, 0.5, 2]      # shape: [3]
U = [[1., 2],
     [-1, 1],
     [2, -0.5]]        # shape: [3, 2]
m = [4., 5]            # shape: [2]
mvn = tfd.MultivariateNormalDiagPlusLowRank(
    loc=mu
    scale_diag=d
    scale_perturb_factor=U,
    scale_perturb_diag=m)

# Evaluate this on an observation in `R^3`, returning a scalar.
mvn.prob([-1, 0, 1]).eval()  # shape: []

# Initialize a 2-batch of 3-variate Gaussians; `S = diag(d) + U @ U.T`.
mu = [[1.,  2,  3],
      [11, 22, 33]]      # shape: [b, k] = [2, 3]
U = [[[1., 2],
      [3,  4],
      [5,  6]],
     [[0.5, 0.75],
      [1,0, 0.25],
      [1.5, 1.25]]]      # shape: [b, k, r] = [2, 3, 2]
m = [[0.1, 0.2],
     [0.4, 0.5]]         # shape: [b, r] = [2, 2]

mvn = tfd.MultivariateNormalDiagPlusLowRank(
    loc=mu,
    scale_perturb_factor=U,
    scale_perturb_diag=m)

mvn.covariance().eval()   # shape: [2, 3, 3]
# ==> [[[  15.63   31.57    48.51]
#       [  31.57   69.31   105.05]
#       [  48.51  105.05   162.59]]
#
#      [[   2.59    1.41    3.35]
#       [   1.41    2.71    3.34]
#       [   3.35    3.34    8.35]]]

# Compute the pdf of two `R^3` observations (one from each batch);
# return a length-2 vector.
x = [[-0.9, 0, 0.1],
     [-10, 0, 9]]     # shape: [2, 3]
mvn.prob(x).eval()    # shape: [2]

loc Floating-point Tensor. If this is set to None, loc is implicitly 0. When specified, must have shape [B1, ..., Bb, k] where b >= 0 and k is the event size.
scale_diag Floating-point Tensor representing a non-singular diagonal matrix added to scale. Must have shape [B1, ..., Bb, k], b >= 0, and characterizes b-batches of k x k diagonal matrices added to scale. When scale_diag is None it defaults to the Identity matrix.
scale_perturb_factor Floating-point Tensor representing a rank-r perturbation added to scale. Must have shape [B1, ..., Bb, k, r], b >= 0, and characterizes b-batches of rank-r updates to scale. When None, no rank-r update is added to scale.
scale_perturb_diag Floating-point Tensor representing a non-singular diagonal matrix inside the rank-r perturbation added to scale. Must have shape [B1, ..., Bb, r], b >= 0, and characterizes b-batches of r x r diagonal matrices inside the perturbation added to scale. When None, an identity matrix is used inside the perturbation. Can only be specified if scale_perturb_factor is also specified.
validate_args Python bool, default False. When True distribution parameters are checked for validity despite possibly degrading runtime performance. When False invalid inputs may silently render incorrect outputs.
allow_nan_stats Python bool, default True. When True, statistics (e.g., mean, mode, variance) use the value 'NaN' to indicate the result is undefined. When False, an exception is raised if one or more of the statistic's batch members are undefined.
name Python str name prefixed to Ops created by this class.

ValueError if at most scale_identity_multiplier is specified.

allow_nan_stats Python bool describing behavior when a stat is undefined.

Stats return +/- infinity when it makes sense. E.g., the variance of a Cauchy distribution is infinity. However, sometimes the statistic is undefined, e.g., if a distribution's pdf does not achieve a maximum within the support of the distribution, the mode is undefined. If the mean is undefined, then by definition the variance is undefined. E.g. the mean for Student's T for df = 1 is undefined (no clear way to say it is either + or - infinity), so the variance = E[(X - mean)**2] is also undefined.

batch_shape Shape of a single sample from a single event index as a TensorShape.

May be partially defined or unknown.

The batch dimensions are indexes into independent, non-identical parameterizations of this distribution.

bijector Function transforming x => y.
distribution Base distribution, p(x).
dtype The DType of Tensors handled by this Distribution.
event_shape Shape of a single sample from a single batch as a TensorShape.

May be partially defined or unknown.

loc The loc Tensor in Y = scale @ X + loc.
name Name prepended to all ops created by this Distribution.
parameters Dictionary of parameters used to instantiate this Distribution.
reparameterization_type Describes how samples from the distribution are reparameterized.

Currently this is one of the static instances tfd.FULLY_REPARAMETERIZED or tfd.NOT_REPARAMETERIZED.

scale The scale LinearOperator in Y = scale @ X + loc.
trainable_variables

validate_args Python bool indicating possibly expensive checks are enabled.
variables

Methods

batch_shape_tensor

View source

Shape of a single sample from a single event index as a 1-D Tensor.

The batch dimensions are indexes into independent, non-identical parameterizations of this distribution.

Args
name name to give to the op

Returns
batch_shape Tensor.

cdf

View source

Cumulative distribution function.

Given random variable X, the cumulative distribution function cdf is:

cdf(x) := P[X <= x]

Args
value float or double Tensor.
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
cdf a Tensor of shape sample_shape(x) + self.batch_shape with values of type self.dtype.

copy

View source

Creates a deep copy of the distribution.

Args
**override_parameters_kwargs String/value dictionary of initialization arguments to override with new values.

Returns
distribution A new instance of type(self) initialized from the union of self.parameters and override_parameters_kwargs, i.e., dict(self.parameters, **override_parameters_kwargs).

covariance

View source

Covariance.

Covariance is (possibly) defined only for non-scalar-event distributions.

For example, for a length-k, vector-valued distribution, it is calculated as,

Cov[i, j] = Covariance(X_i, X_j) = E[(X_i - E[X_i]) (X_j - E[X_j])]

where Cov is a (batch of) k x k matrix, 0 <= (i, j) < k, and E denotes expectation.

Alternatively, for non-vector, multivariate distributions (e.g., matrix-valued, Wishart), Covariance shall return a (batch of) matrices under some vectorization of the events, i.e.,

Cov[i, j] = Covariance(Vec(X)_i, Vec(X)_j) = [as above]

where Cov is a (batch of) k' x k' matrices, 0 <= (i, j) < k' = reduce_prod(event_shape), and Vec is some function mapping indices of this distribution's event dimensions to indices of a length-k' vector.

Args
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
covariance Floating-point Tensor with shape [B1, ..., Bn, k', k'] where the first n dimensions are batch coordinates and k' = reduce_prod(self.event_shape).

cross_entropy

View source

Computes the (Shannon) cross entropy.

Denote this distribution (self) by P and the other distribution by Q. Assuming P, Q are absolutely continuous with respect to one another and permit densities p(x) dr(x) and q(x) dr(x), (Shannon) cross entropy is defined as:

H[P, Q] = E_p[-log q(X)] = -int_F p(x) log q(x) dr(x)

where F denotes the support of the random variable X ~ P.

other types with built-in registrations: GaussianProcess, GaussianProcessRegressionModel, MultivariateNormalDiag, MultivariateNormalDiagPlusLowRank, MultivariateNormalFullCovariance, MultivariateNormalLinearOperator, MultivariateNormalTriL, VariationalGaussianProcess

Args
other tfp.distributions.Distribution instance.
name Python str prepended to names of ops created by this function.

Returns
cross_entropy self.dtype Tensor with shape [B1, ..., Bn] representing n different calculations of (Shannon) cross entropy.

entropy

View source

Shannon entropy in nats.

event_shape_tensor

View source

Shape of a single sample from a single batch as a 1-D int32 Tensor.

Args
name name to give to the op

Returns
event_shape Tensor.

is_scalar_batch

View source

Indicates that batch_shape == [].

Args
name Python str prepended to names of ops created by this function.

Returns
is_scalar_batch bool scalar Tensor.

is_scalar_event

View source

Indicates that event_shape == [].

Args
name Python str prepended to names of ops created by this function.

Returns
is_scalar_event bool scalar Tensor.

kl_divergence

View source

Computes the Kullback--Leibler divergence.

Denote this distribution (self) by p and the other distribution by q. Assuming p, q are absolutely continuous with respect to reference measure r, the KL divergence is defined as:

KL[p, q] = E_p[log(p(X)/q(X))]
         = -int_F p(x) log q(x) dr(x) + int_F p(x) log p(x) dr(x)
         = H[p, q] - H[p]

where F denotes the support of the random variable X ~ p, H[., .] denotes (Shannon) cross entropy, and H[.] denotes (Shannon) entropy.

other types with built-in registrations: GaussianProcess, GaussianProcessRegressionModel, MultivariateNormalDiag, MultivariateNormalDiagPlusLowRank, MultivariateNormalFullCovariance, MultivariateNormalLinearOperator, MultivariateNormalTriL, VariationalGaussianProcess

Args
other tfp.distributions.Distribution instance.
name Python str prepended to names of ops created by this function.

Returns
kl_divergence self.dtype Tensor with shape [B1, ..., Bn] representing n different calculations of the Kullback-Leibler divergence.

log_cdf

View source

Log cumulative distribution function.

Given random variable X, the cumulative distribution function cdf is:

log_cdf(x) := Log[ P[X <= x] ]

Often, a numerical approximation can be used for log_cdf(x) that yields a more accurate answer than simply taking the logarithm of the cdf when x << -1.

Args
value float or double Tensor.
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
logcdf a Tensor of shape sample_shape(x) + self.batch_shape with values of type self.dtype.

log_prob

View source

Log probability density/mass function.

Additional documentation from MultivariateNormalLinearOperator:

value is a batch vector with compatible shape if value is a Tensor whose shape can be broadcast up to either:

self.batch_shape + self.event_shape

or

[M1, ..., Mm] + self.batch_shape + self.event_shape

Args
value float or double Tensor.
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
log_prob a Tensor of shape sample_shape(x) + self.batch_shape with values of type self.dtype.

log_survival_function

View source

Log survival function.

Given random variable X, the survival function is defined:

log_survival_function(x) = Log[ P[X > x] ]
                         = Log[ 1 - P[X <= x] ]
                         = Log[ 1 - cdf(x) ]

Typically, different numerical approximations can be used for the log survival function, which are more accurate than 1 - cdf(x) when x >> 1.

Args
value float or double Tensor.
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
Tensor of shape sample_shape(x) + self.batch_shape with values of type self.dtype.

mean

View source

Mean.

mode

View source

Mode.

param_shapes

View source

Shapes of parameters given the desired shape of a call to sample().

This is a class method that describes what key/value arguments are required to instantiate the given Distribution so that a particular shape is returned for that instance's call to sample().

Subclasses should override class method _param_shapes.

Args
sample_shape Tensor or python list/tuple. Desired shape of a call to sample().
name name to prepend ops with.

Returns
dict of parameter name to Tensor shapes.

param_static_shapes

View source

param_shapes with static (i.e. TensorShape) shapes.

This is a class method that describes what key/value arguments are required to instantiate the given Distribution so that a particular shape is returned for that instance's call to sample(). Assumes that the sample's shape is known statically.

Subclasses should override class method _param_shapes to return constant-valued tensors when constant values are fed.

Args
sample_shape TensorShape or python list/tuple. Desired shape of a call to sample().

Returns
dict of parameter name to TensorShape.

Raises
ValueError if sample_shape is a TensorShape and is not fully defined.

prob

View source

Probability density/mass function.

Additional documentation from MultivariateNormalLinearOperator:

value is a batch vector with compatible shape if value is a Tensor whose shape can be broadcast up to either:

self.batch_shape + self.event_shape

or

[M1, ..., Mm] + self.batch_shape + self.event_shape

Args
value float or double Tensor.
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
prob a Tensor of shape sample_shape(x) + self.batch_shape with values of type self.dtype.

quantile

View source

Quantile function. Aka 'inverse cdf' or 'percent point function'.

Given random variable X and p in [0, 1], the quantile is:

quantile(p) := x such that P[X <= x] == p

Args
value float or double Tensor.
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
quantile a Tensor of shape sample_shape(x) + self.batch_shape with values of type self.dtype.

sample

View source

Generate samples of the specified shape.

Note that a call to sample() without arguments will generate a single sample.

Args
sample_shape 0D or 1D int32 Tensor. Shape of the generated samples.
seed Python integer or tfp.util.SeedStream instance, for seeding PRNG.
name name to give to the op.
**kwargs Named arguments forwarded to subclass implementation.

Returns
samples a Tensor with prepended dimensions sample_shape.

stddev

View source

Standard deviation.

Standard deviation is defined as,

stddev = E[(X - E[X])**2]**0.5

where X is the random variable associated with this distribution, E denotes expectation, and stddev.shape = batch_shape + event_shape.

Args
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
stddev Floating-point Tensor with shape identical to batch_shape + event_shape, i.e., the same shape as self.mean().

survival_function

View source

Survival function.

Given random variable X, the survival function is defined:

survival_function(x) = P[X > x]
                     = 1 - P[X <= x]
                     = 1 - cdf(x).

Args
value float or double Tensor.
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
Tensor of shape sample_shape(x) + self.batch_shape with values of type self.dtype.

variance

View source

Variance.

Variance is defined as,

Var = E[(X - E[X])**2]

where X is the random variable associated with this distribution, E denotes expectation, and Var.shape = batch_shape + event_shape.

Args
name Python str prepended to names of ops created by this function.
**kwargs Named arguments forwarded to subclass implementation.

Returns
variance Floating-point Tensor with shape identical to batch_shape + event_shape, i.e., the same shape as self.mean().

__getitem__

View source

Slices the batch axes of this distribution, returning a new instance.

b = tfd.Bernoulli(logits=tf.zeros([3, 5, 7, 9]))
b.batch_shape  # => [3, 5, 7, 9]
b2 = b[:, tf.newaxis, ..., -2:, 1::2]
b2.batch_shape  # => [3, 1, 5, 2, 4]

x = tf.random.stateless_normal([5, 3, 2, 2])
cov = tf.matmul(x, x, transpose_b=True)
chol = tf.linalg.cholesky(cov)
loc = tf.random.stateless_normal([4, 1, 3, 1])
mvn = tfd.MultivariateNormalTriL(loc, chol)
mvn.batch_shape  # => [4, 5, 3]
mvn.event_shape  # => [2]
mvn2 = mvn[:, 3:, ..., ::-1, tf.newaxis]
mvn2.batch_shape  # => [4, 2, 3, 1]
mvn2.event_shape  # => [2]

Args
slices slices from the [] operator

Returns
dist A new tfd.Distribution instance with sliced parameters.

__iter__

View source