Scale: Number of records in the input dataset.
Shape: aka Empirical Distribution. The proportions of the counts that reside in each bin.
Domain Size: Number of bins the domain is divided into.
Input Dataset: The input is represented as a histogram of counts. The domain is divided into bins of equal width.
Output Dataset: The output is also represented as a histogram of counts over a uniform grid. The resolution of the grid matches that of the input. While the algorithms themselves may not actually generate a histogram, our visualization represents the histogram inferred from the noisy counts generated by the algorithm.
Epsilon: The privacy parameter. Controls how much information is disclosed about each record in the input dataset by a differentially private algorithm.
Algorithm: A differentially private algorithm. We consider algorithms that permit answering 1- and 2-dimensional range queries under differential privacy.
Data Independent: We call an algorithm data independent if the absolute error incurred by the algorithm is independent of the input dataset's shape and scale.