Average performance leadership : With an average precision of 0.6925 across all datasets, AVE outperforms the second-best method (DAGMM at 0.5496) by 25.99% which is a highly significant margin in precision metrics where differences of 1–2% are typically considered significant.
← all excerpts
A new hybrid model for improving outlier detection using combined autoencoder and variational autoencoder.
1
—
—