Sorting Data via a Look-Up-Table Neural Network and Self-Regulating Index

Figure 9

Illustration of some experimental results. In all charts, the x-axis represents the values of unsorted elements in the test dataset and the y-axis represents their CDF values. All charts are drawn based on the same dataset in normal distribution with a 1 million scale. The pink triangle-shaped points depict the CDF values predicted by predictive models, the blue points present the true CDF values of unsorted elements in test datasets, and the lower and upper gray lines are the lower and upper bounds of the CDF (i.e., 0 and 1), respectively. (a) LNN model, (b) LR model, (c) Ada Boost model, and (d) DNN model.

(a)

(b)

(c)

(d)