likelihoods.py
# Copyright 2016 Valentine Svensson, James Hensman, alexggmatthews, Alexis Boukouvalas
# Copyright 2017 Artem Artemev @awav
#
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#
# Unless required by applicable law or agreed to in writing, software
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
r"""
Likelihoods are another core component of GPflow. This describes how likely the
p(Y|F). Different likelihoods make different
assumptions about the distribution of the data, as such different data-types
(continuous, binary, ordinal, count) are better modelled with different
likelihood assumptions.

Use of any likelihood other than Gaussian typically introduces the need to use
an approximation to perform inference, if one isn't already needed. A
variational inference and MCMC models are included in GPflow and allow
approximate inference with non-Gaussian likelihoods. An introduction to these
models can be found :ref:here <implemented_models>. Specific notebooks
illustrating non-Gaussian likelihood regressions are available for
classification <notebooks/classification.html>_ (binary data), ordinal
<notebooks/ordinal.html>_ and multiclass <notebooks/multiclass.html>_.

Creating new likelihoods
----------
Likelihoods are defined by their
log-likelihood. When creating new likelihoods, the
:func:logp <gpflow.likelihoods.Likelihood.logp> method (log p(Y|F)), the
:func:conditional_mean <gpflow.likelihoods.Likelihood.conditional_mean>,
:func:conditional_variance
<gpflow.likelihoods.Likelihood.conditional_variance>.

In order to perform variational inference with non-Gaussian likelihoods a term
called variational expectations, ∫ q(F) log p(Y|F) dF, needs to
be computed under a Gaussian distribution q(F) ~ N(μ, Σ).

The :func:variational_expectations <gpflow.likelihoods.Likelihood.variational_expectations>
method can be overriden if this can be computed in closed form, otherwise; if
the new likelihood inherits
:class:Likelihood <gpflow.likelihoods.Likelihood> the default will use
Gauss-Hermite numerical integration (works well when F is 1D
or 2D), if the new likelihood inherits from
:class:MonteCarloLikelihood <gpflow.likelihoods.MonteCarloLikelihood> the
integration is done by sampling (can be more suitable when F is higher dimensional).
"""

import numpy as np
import tensorflow as tf

from . import logdensities
from . import priors
from . import settings
from . import transforms
from .decors import params_as_tensors
from .decors import params_as_tensors_for
from .params import ParamList
from .params import Parameter
from .params import Parameterized

class Likelihood(Parameterized):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.num_gauss_hermite_points = 20

def predict_mean_and_var(self, Fmu, Fvar):
r"""
Given a Normal distribution for the latent function,
return the mean of Y

if
q(f) = N(Fmu, Fvar)

and this object represents

p(y|f)

then this method computes the predictive mean

\int\int y p(y|f)q(f) df dy

and the predictive variance

\int\int y^2 p(y|f)q(f) df dy  - [ \int\int y p(y|f)q(f) df dy ]^2

Here, we implement a default Gauss-Hermite quadrature routine, but some
likelihoods (e.g. Gaussian) will implement specific cases.
"""
integrand2 = lambda *X: self.conditional_variance(*X) + tf.square(self.conditional_mean(*X))
self.num_gauss_hermite_points,
Fmu, Fvar)
V_y = E_y2 - tf.square(E_y)
return E_y, V_y

def predict_density(self, Fmu, Fvar, Y):
r"""
Given a Normal distribution for the latent function, and a datum Y,
compute the log predictive density of Y.

i.e. if
q(f) = N(Fmu, Fvar)

and this object represents

p(y|f)

then this method computes the predictive density

\log \int p(y=Y|f)q(f) df

Here, we implement a default Gauss-Hermite quadrature routine, but some
likelihoods (Gaussian, Poisson) will implement specific cases.
"""
self.num_gauss_hermite_points,
Fmu, Fvar, logspace=True, Y=Y)

def variational_expectations(self, Fmu, Fvar, Y):
r"""
Compute the expected log density of the data, given a Gaussian
distribution for the function values.

if
q(f) = N(Fmu, Fvar)

and this object represents

p(y|f)

then this method computes

\int (\log p(y|f)) q(f) df.

Here, we implement a default Gauss-Hermite quadrature routine, but some
likelihoods (Gaussian, Poisson) will implement specific cases.
"""
self.num_gauss_hermite_points,
Fmu, Fvar, Y=Y)

class Gaussian(Likelihood):
def __init__(self, variance=1.0, **kwargs):
super().__init__(**kwargs)
self.variance = Parameter(
variance, transform=transforms.positive, dtype=settings.float_type)

@params_as_tensors
def logp(self, F, Y):
return logdensities.gaussian(Y, F, self.variance)

@params_as_tensors
def conditional_mean(self, F):  # pylint: disable=R0201
return tf.identity(F)

@params_as_tensors
def conditional_variance(self, F):
return tf.fill(tf.shape(F), tf.squeeze(self.variance))

@params_as_tensors
def predict_mean_and_var(self, Fmu, Fvar):
return tf.identity(Fmu), Fvar + self.variance

@params_as_tensors
def predict_density(self, Fmu, Fvar, Y):
return logdensities.gaussian(Y, Fmu, Fvar + self.variance)

@params_as_tensors
def variational_expectations(self, Fmu, Fvar, Y):
return -0.5 * np.log(2 * np.pi) - 0.5 * tf.log(self.variance) \
- 0.5 * (tf.square(Y - Fmu) + Fvar) / self.variance

class Poisson(Likelihood):
"""
Poisson likelihood for use with count data, where the rate is given by the (transformed) GP.

let g(.) be the inverse-link function, then this likelihood represents

p(y_i | f_i) = Poisson(y_i | g(f_i) * binsize)

Note:binsize
For use in a Log Gaussian Cox process (doubly stochastic model) where the
rate function of an inhomogeneous Poisson process is given by a GP.  The
intractable likelihood can be approximated by gridding the space (into bins
of size 'binsize') and using this Poisson likelihood.
"""

super().__init__(**kwargs)
self.binsize = np.double(binsize)

def logp(self, F, Y):

def conditional_variance(self, F):

def conditional_mean(self, F):

def variational_expectations(self, Fmu, Fvar, Y):
return Y * Fmu - tf.exp(Fmu + Fvar / 2) * self.binsize \
- tf.lgamma(Y + 1) + Y * tf.log(self.binsize)
return super(Poisson, self).variational_expectations(Fmu, Fvar, Y)

class Exponential(Likelihood):
super().__init__(**kwargs)

def logp(self, F, Y):

def conditional_mean(self, F):

def conditional_variance(self, F):

def variational_expectations(self, Fmu, Fvar, Y):
return - tf.exp(-Fmu + Fvar / 2) * Y - Fmu
return super().variational_expectations(Fmu, Fvar, Y)

class StudentT(Likelihood):
def __init__(self, scale=1.0, df=3.0, **kwargs):
"""
:param scale float: scale parameter
:param df float: degrees of freedom
"""
super().__init__(**kwargs)
self.df = df
self.scale = Parameter(scale, transform=transforms.positive,
dtype=settings.float_type)

@params_as_tensors
def logp(self, F, Y):
return logdensities.student_t(Y, F, self.scale, self.df)

@params_as_tensors
def conditional_mean(self, F):
return tf.identity(F)

@params_as_tensors
def conditional_variance(self, F):
var = self.scale ** 2 * (self.df / (self.df - 2.0))
return tf.fill(tf.shape(F), tf.squeeze(var))

def inv_probit(x):
jitter = 1e-3  # ensures output is strictly between 0 and 1
return 0.5 * (1.0 + tf.erf(x / np.sqrt(2.0))) * (1 - 2 * jitter) + jitter

class Bernoulli(Likelihood):
super().__init__(**kwargs)

def logp(self, F, Y):

def predict_mean_and_var(self, Fmu, Fvar):
p = inv_probit(Fmu / tf.sqrt(1 + Fvar))
return p, p - tf.square(p)
else:
return super().predict_mean_and_var(Fmu, Fvar)

def predict_density(self, Fmu, Fvar, Y):
p = self.predict_mean_and_var(Fmu, Fvar)[0]
return logdensities.bernoulli(Y, p)

def conditional_mean(self, F):

def conditional_variance(self, F):
p = self.conditional_mean(F)
return p - tf.square(p)

class Gamma(Likelihood):
"""
Use the transformed GP to give the *scale* (inverse rate) of the Gamma
"""

super().__init__(**kwargs)
self.shape = Parameter(1.0, transform=transforms.positive)

@params_as_tensors
def logp(self, F, Y):

@params_as_tensors
def conditional_mean(self, F):

@params_as_tensors
def conditional_variance(self, F):
return self.shape * tf.square(scale)

@params_as_tensors
def variational_expectations(self, Fmu, Fvar, Y):
return -self.shape * Fmu - tf.lgamma(self.shape) \
+ (self.shape - 1.) * tf.log(Y) - Y * tf.exp(-Fmu + Fvar / 2.)
else:
return super().variational_expectations(Fmu, Fvar, Y)

class Beta(Likelihood):
"""
This uses a reparameterisation of the Beta density. We have the mean of the
Beta distribution given by the transformed process:

m = sigma(f)

and a scale parameter. The familiar alpha, beta parameters are given by

m     = alpha / (alpha + beta)
scale = alpha + beta

so:
alpha = scale * m
beta  = scale * (1-m)
"""

super().__init__(**kwargs)
self.scale = Parameter(scale, transform=transforms.positive)

@params_as_tensors
def logp(self, F, Y):
alpha = mean * self.scale
beta = self.scale - alpha
return logdensities.beta(Y, alpha, beta)

@params_as_tensors
def conditional_mean(self, F):

@params_as_tensors
def conditional_variance(self, F):
return (mean - tf.square(mean)) / (self.scale + 1.)

class RobustMax(Parameterized):
"""
This class represent a multi-class inverse-link function. Given a vector
f=[f_1, f_2, ... f_k], the result of the mapping is

y = [y_1 ... y_k]

with

y_i = (1-eps)  i == argmax(f)
eps/(k-1)  otherwise.
"""

def __init__(self, num_classes, epsilon=1e-3, **kwargs):
super().__init__(**kwargs)
self.epsilon = Parameter(epsilon, transforms.Logistic(), trainable=False, dtype=settings.float_type,
prior=priors.Beta(0.2, 5.))
self.num_classes = num_classes

@params_as_tensors
def __call__(self, F):
i = tf.argmax(F, 1)
return tf.one_hot(i, self.num_classes, tf.squeeze(1. - self.epsilon), tf.squeeze(self._eps_K1))

@property
@params_as_tensors
def _eps_K1(self):
return self.epsilon / (self.num_classes - 1.)

def prob_is_largest(self, Y, mu, var, gh_x, gh_w):
Y = tf.cast(Y, tf.int64)
# work out what the mean and variance is of the indicated latent function.
oh_on = tf.cast(tf.one_hot(tf.reshape(Y, (-1,)), self.num_classes, 1., 0.), settings.float_type)
mu_selected = tf.reduce_sum(oh_on * mu, 1)
var_selected = tf.reduce_sum(oh_on * var, 1)

# generate Gauss Hermite grid
X = tf.reshape(mu_selected, (-1, 1)) + gh_x * tf.reshape(
tf.sqrt(tf.clip_by_value(2. * var_selected, 1e-10, np.inf)), (-1, 1))

# compute the CDF of the Gaussian between the latent functions and the grid (including the selected function)
dist = (tf.expand_dims(X, 1) - tf.expand_dims(mu, 2)) / tf.expand_dims(
tf.sqrt(tf.clip_by_value(var, 1e-10, np.inf)), 2)
cdfs = 0.5 * (1.0 + tf.erf(dist / np.sqrt(2.0)))

cdfs = cdfs * (1 - 2e-4) + 1e-4

# blank out all the distances on the selected latent function
oh_off = tf.cast(tf.one_hot(tf.reshape(Y, (-1,)), self.num_classes, 0., 1.), settings.float_type)
cdfs = cdfs * tf.expand_dims(oh_off, 2) + tf.expand_dims(oh_on, 2)

# take the product over the latent functions, and the sum over the GH grid.
return tf.matmul(tf.reduce_prod(cdfs, reduction_indices=[1]), tf.reshape(gh_w / np.sqrt(np.pi), (-1, 1)))

class MultiClass(Likelihood):
"""
A likelihood that can do multi-way classification.
Currently the only valid choice
"""
super().__init__(**kwargs)
self.num_classes = num_classes
raise NotImplementedError

def logp(self, F, Y):
hits = tf.equal(tf.expand_dims(tf.argmax(F, 1), 1), tf.cast(Y, tf.int64))
yes = tf.ones(tf.shape(Y), dtype=settings.float_type) - self.invlink.epsilon
no = tf.zeros(tf.shape(Y), dtype=settings.float_type) + self.invlink._eps_K1
p = tf.where(hits, yes, no)
return tf.log(p)
else:
raise NotImplementedError

def variational_expectations(self, Fmu, Fvar, Y):
gh_x, gh_w = hermgauss(self.num_gauss_hermite_points)
p = self.invlink.prob_is_largest(Y, Fmu, Fvar, gh_x, gh_w)
ve = p * tf.log(1. - self.invlink.epsilon) + (1. - p) * tf.log(self.invlink._eps_K1)
return ve
else:
raise NotImplementedError

def predict_mean_and_var(self, Fmu, Fvar):
# To compute this, we'll compute the density for each possible output
possible_outputs = [tf.fill(tf.stack([tf.shape(Fmu)[0], 1]), np.array(i, dtype=np.int64)) for i in
range(self.num_classes)]
ps = [self._predict_non_logged_density(Fmu, Fvar, po) for po in possible_outputs]
ps = tf.transpose(tf.stack([tf.reshape(p, (-1,)) for p in ps]))
return ps, ps - tf.square(ps)
else:
raise NotImplementedError

def predict_density(self, Fmu, Fvar, Y):
return tf.log(self._predict_non_logged_density(Fmu, Fvar, Y))

def _predict_non_logged_density(self, Fmu, Fvar, Y):
gh_x, gh_w = hermgauss(self.num_gauss_hermite_points)
p = self.invlink.prob_is_largest(Y, Fmu, Fvar, gh_x, gh_w)
den = p * (1. - self.invlink.epsilon) + (1. - p) * (self.invlink._eps_K1)
return den
else:
raise NotImplementedError

def conditional_mean(self, F):

def conditional_variance(self, F):
p = self.conditional_mean(F)
return p - tf.square(p)

class SwitchedLikelihood(Likelihood):
def __init__(self, likelihood_list, **kwargs):
"""
In this likelihood, we assume at extra column of Y, which contains
integers that specify a likelihood from the list of likelihoods.
"""
super().__init__(**kwargs)
for l in likelihood_list:
assert isinstance(l, Likelihood)
self.likelihood_list = ParamList(likelihood_list)
self.num_likelihoods = len(self.likelihood_list)

def _partition_and_stitch(self, args, func_name):
"""
args is a list of tensors, to be passed to self.likelihoods.<func_name>

args[-1] is the 'Y' argument, which contains the indexes to self.likelihoods.

This function splits up the args using dynamic_partition, calls the
relevant function on the likelihoods, and re-combines the result.
"""
# get the index from Y
Y = args[-1]
ind = Y[:, -1]
ind = tf.cast(ind, tf.int32)
Y = Y[:, :-1]
args[-1] = Y

# split up the arguments into chunks corresponding to the relevant likelihoods
args = zip(*[tf.dynamic_partition(X, ind, self.num_likelihoods) for X in args])

# apply the likelihood-function to each section of the data
with params_as_tensors_for(self, convert=False):
funcs = [getattr(lik, func_name) for lik in self.likelihood_list]
results = [f(*args_i) for f, args_i in zip(funcs, args)]

# stitch the results back together
partitions = tf.dynamic_partition(tf.range(0, tf.size(ind)), ind, self.num_likelihoods)
results = tf.dynamic_stitch(partitions, results)

return results

def logp(self, F, Y):
return self._partition_and_stitch([F, Y], 'logp')

def predict_density(self, Fmu, Fvar, Y):
return self._partition_and_stitch([Fmu, Fvar, Y], 'predict_density')

def variational_expectations(self, Fmu, Fvar, Y):
return self._partition_and_stitch([Fmu, Fvar, Y], 'variational_expectations')

def predict_mean_and_var(self, Fmu, Fvar):
mvs = [lik.predict_mean_and_var(Fmu, Fvar) for lik in self.likelihood_list]
mu_list, var_list = zip(*mvs)
mu = tf.concat(mu_list, 1)
var = tf.concat(var_list, 1)
return mu, var

class Ordinal(Likelihood):
"""
A likelihood for doing ordinal regression.

The data are integer values from 0 to K, and the user must specify (K-1)
'bin edges' which define the points at which the labels switch. Let the bin
edges be [a_0, a_1, ... a_{K-1}], then the likelihood is

p(Y=0|F) = phi((a_0 - F) / sigma)
p(Y=1|F) = phi((a_1 - F) / sigma) - phi((a_0 - F) / sigma)
p(Y=2|F) = phi((a_2 - F) / sigma) - phi((a_1 - F) / sigma)
...
p(Y=K|F) = 1 - phi((a_{K-1} - F) / sigma)

where phi is the cumulative density function of a Gaussian (the inverse probit
function) and sigma is a parameter to be learned. A reference is:

@article{chu2005gaussian,
title={Gaussian processes for ordinal regression},
author={Chu, Wei and Ghahramani, Zoubin},
journal={Journal of Machine Learning Research},
volume={6},
number={Jul},
pages={1019--1041},
year={2005}
}
"""

def __init__(self, bin_edges, **kwargs):
"""
bin_edges is a numpy array specifying at which function value the
output label should switch. If the possible Y values are 0...K, then
the size of bin_edges should be (K-1).
"""
super().__init__(**kwargs)
self.bin_edges = bin_edges
self.num_bins = bin_edges.size + 1
self.sigma = Parameter(1.0, transform=transforms.positive)

@params_as_tensors
def logp(self, F, Y):
Y = tf.cast(Y, tf.int64)
scaled_bins_left = tf.concat([self.bin_edges / self.sigma, np.array([np.inf])], 0)
scaled_bins_right = tf.concat([np.array([-np.inf]), self.bin_edges / self.sigma], 0)
selected_bins_left = tf.gather(scaled_bins_left, Y)
selected_bins_right = tf.gather(scaled_bins_right, Y)

return tf.log(inv_probit(selected_bins_left - F / self.sigma) -
inv_probit(selected_bins_right - F / self.sigma) + 1e-6)

@params_as_tensors
def _make_phi(self, F):
"""
A helper function for making predictions. Constructs a probability
matrix where each row output the probability of the corresponding
label, and the rows match the entries of F.

Note that a matrix of F values is flattened.
"""
scaled_bins_left = tf.concat([self.bin_edges / self.sigma, np.array([np.inf])], 0)
scaled_bins_right = tf.concat([np.array([-np.inf]), self.bin_edges / self.sigma], 0)
return inv_probit(scaled_bins_left - tf.reshape(F, (-1, 1)) / self.sigma) \
- inv_probit(scaled_bins_right - tf.reshape(F, (-1, 1)) / self.sigma)

def conditional_mean(self, F):
phi = self._make_phi(F)
Ys = tf.reshape(np.arange(self.num_bins, dtype=np.float64), (-1, 1))
return tf.reshape(tf.matmul(phi, Ys), tf.shape(F))

def conditional_variance(self, F):
phi = self._make_phi(F)
Ys = tf.reshape(np.arange(self.num_bins, dtype=np.float64), (-1, 1))
E_y = tf.matmul(phi, Ys)
E_y2 = tf.matmul(phi, tf.square(Ys))
return tf.reshape(E_y2 - tf.square(E_y), tf.shape(F))

class MonteCarloLikelihood(Likelihood):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.num_monte_carlo_points = 100
del self.num_gauss_hermite_points

def _mc_quadrature(self, funcs, Fmu, Fvar, logspace: bool = False, epsilon=None, **Ys):
return ndiag_mc(funcs, self.num_monte_carlo_points, Fmu, Fvar, logspace, epsilon, **Ys)

def predict_mean_and_var(self, Fmu, Fvar, epsilon=None):
r"""
Given a Normal distribution for the latent function,
return the mean of Y

if
q(f) = N(Fmu, Fvar)

and this object represents

p(y|f)

then this method computes the predictive mean

\int\int y p(y|f)q(f) df dy

and the predictive variance

\int\int y^2 p(y|f)q(f) df dy  - [ \int\int y p(y|f)q(f) df dy ]^2

Here, we implement a default Monte Carlo routine.
"""
integrand2 = lambda *X: self.conditional_variance(*X) + tf.square(self.conditional_mean(*X))
Fmu, Fvar, epsilon=epsilon)
V_y = E_y2 - tf.square(E_y)
return E_y, V_y  # N x D

def predict_density(self, Fmu, Fvar, Y, epsilon=None):
r"""
Given a Normal distribution for the latent function, and a datum Y,
compute the log predictive density of Y.

i.e. if
q(f) = N(Fmu, Fvar)

and this object represents

p(y|f)

then this method computes the predictive density

\log \int p(y=Y|f)q(f) df

Here, we implement a default Monte Carlo routine.
"""
return self._mc_quadrature(self.logp, Fmu, Fvar, Y=Y, logspace=True, epsilon=epsilon)

def variational_expectations(self, Fmu, Fvar, Y, epsilon=None):
r"""
Compute the expected log density of the data, given a Gaussian
distribution for the function values.

if
q(f) = N(Fmu, Fvar)  - Fmu: N x D  Fvar: N x D

and this object represents

p(y|f)  - Y: N x 1

then this method computes

\int (\log p(y|f)) q(f) df.

Here, we implement a default Monte Carlo quadrature routine.
"""
return self._mc_quadrature(self.logp, Fmu, Fvar, Y=Y, epsilon=epsilon)

class GaussianMC(MonteCarloLikelihood, Gaussian):
"""
Stochastic version of Gaussian likelihood for comparison.
"""
pass

class SoftMax(MonteCarloLikelihood):
"""
The soft-max multi-class likelihood.
"""

def __init__(self, num_classes, **kwargs):
super().__init__(**kwargs)
self.num_classes = num_classes

def logp(self, F, Y):
with tf.control_dependencies(
[
tf.assert_equal(tf.shape(Y)[1], 1),
tf.assert_equal(tf.cast(tf.shape(F)[1], settings.int_type),
tf.cast(self.num_classes, settings.int_type))
]):
if Y.dtype != np.int32:
Y = tf.cast(Y, np.int32)
return -tf.nn.sparse_softmax_cross_entropy_with_logits(logits=F, labels=Y[:, 0])[:, None]

def conditional_mean(self, F):
return tf.nn.softmax(F)

def conditional_variance(self, F):
p = self.conditional_mean(F)
return p - tf.square(p)