tfp.experimental.bayesopt.acquisition.GaussianProcessExpectedImprovement

View source on GitHub

Gaussian Process expected improvement acquisition function.

Inherits From: AcquisitionFunction

tfp.experimental.bayesopt.acquisition.GaussianProcessExpectedImprovement(
    predictive_distribution, observations, seed=None, exploration=0.01
)

Computes the analytic sequential expected improvement for a Gaussian process model.

Requires that predictive_distribution has a mean, stddev method.

Examples

Build and evaluate a Gausian Process Expected Improvement acquisition function.

import numpy as np
import tensorflow_probability as tfp

tfd = tfp.distributions
tfpk = tfp.math.psd_kernels
tfp_acq = tfp.experimental.bayesopt.acquisition

# Sample 10 20-dimensional index points and associated observations.
index_points = np.random.uniform(size=[10, 20])
observations = np.random.uniform(size=[10])

# Build a Gaussian Process regression model over the function values at
# `predictive_index_points` conditioned on observations.
predictive_index_points = np.random.uniform(size=[8, 20])
dist = tfd.GaussianProcessRegressionModel(
    kernel=tfpk.MaternFiveHalves(),
    observation_index_points=index_points,
    observations=observations,
    predictive_index_points=predictive_index_points)

# Define a GP Expected Improvement acquisition function.
gp_ei = tfp_acq.GaussianProcessExpectedImprovement(
    predictive_distribution=dist,
    observations=observations)

# Evaluate the acquisition function at `predictive_index_points`.
acq_fn_vals = gp_ei()

# Evaluate the acquisition function at a new set of predictive index points.
pred_index_points = np.random.uniform(size=[6, 20])
acq_fn_vals = gp_ei(pred_index_points)

Args
predictive_distribution	tfd.Distribution-like, the distribution over observations at a set of index points. Must have mean, stddev methods.
observations	Float Tensor of observations.
seed	PRNG seed; see tfp.random.sanitize_seed for details.
exploration	Exploitation-exploration trade-off parameter.

Attributes
exploration
is_parallel	Python bool indicating whether the acquisition function is parallel. Parallel (batched) acquisition functions evaluate batches of points rather than single points.
observations	Float Tensor of observations.
predictive_distribution	The distribution over observations at a set of index points.
seed	PRNG seed.

Methods

__call__

View source

__call__(
    **kwargs
)

Computes analytic GP expected improvement.

Args
**kwargs	Keyword args passed on to the mean and stddev methods of predictive_distribution.

Returns
Expected improvements at index points implied by predictive_distribution (or overridden in **kwargs).