boax.acquisitions.upper_confidence_bound

boax.acquisitions.upper_confidence_bound#

boax.acquisitions.upper_confidence_bound(bounds, beta=2.0, num_raw_samples=512, num_restarts=10)#

The Upper Confidence Bound (UCB) acquisition function.

Upper confidence bound comprises of the posterior mean plus an additional term: the posterior standard deviation weighted by a trade-off parameter, beta.

UCB(x) = loc(x) + sqrt(beta) * scale(x)

Example

>>> acqf = upper_confidence_bound(2.0, model)
>>> ucb = acqf(index_points)
Parameters:

  • beta (Union[Array, ndarray, bool, number, float, int]) – The mean and covariance trade-off parameter.

  • model – The surrogate model.

Return type:

Acquisition

Returns:

The corresponding Acquisition.