dodiscover.ci.kernel_utils.corrent_matrix#
- dodiscover.ci.kernel_utils.corrent_matrix(data, metric='rbf', kwidth=None, distance_metric='euclidean', n_jobs=None)[source]#
Compute the centered correntropy of a matrix.
- Parameters:
- dataarray_like of shape (n_samples, n_features)
The data.
- metric
str The kernel metric.
- kwidth
float The kernel width.
- distance_metric
str The distance metric to infer kernel width.
- n_jobs
int, optional The number of jobs to run computations in parallel, by default None.
- Returns:
- dataarray_like of shape (n_features, n_features)
A symmetric centered correntropy matrix of the data.
Notes
The estimator for the correntropy array is given by the formula \(1 / N \sum_{i=1}^N k(x_i, y_i) - 1 / N**2 \sum_{i=1}^N \sum_{j=1}^N k(x_i, y_j)\). The first term is the estimate, and the second term is the bias, and together they form an unbiased estimate.