Table 3 Performance of different methods on different data sets in terms of FDR (top) and TDR (bottom) with nominal level is \(\alpha =0.2\). Mean values and standard deviations (in parentheses) over 100 subsampled data sets
From: Conformal novelty detection for multiple metabolic networks
DD | PROTEINS | AIDS | NCI1 | ||
|---|---|---|---|---|---|
FDR | |||||
GIN | AdaDetect | 0.05 (0.10) | 0.04 (0.12) | 0.10 (0.10) | 0.04 (0.11) |
CAD | 0.04 (0.11) | 0.05 (0.13) | 0.19 (0.08) | 0.04 (0.10) | |
DiffPool | AdaDetect | 0.02 (0.05) | 0.01 (0.05) | 0.06 (0.08) | 0.00 (0.01) |
CAD | 0.00 (0.02) | 0.03 (0.13) | 0.04 (0.08) | 0.00 (0.00) | |
DGCNN | AdaDetect | 0.11 (0.12) | 0.08 (0.12) | 0.19 (0.07) | 0.03 (0.08) |
CAD | 0.03 (0.09) | 0.04 (0.12) | 0.10 (0.10) | 0.03 (0.11) | |
WL | AdaDetect | 0.10 (0.12) | 0.08 (0.12) | 0.19 (0.06) | 0.06 (0.11) |
CAD | 0.08 (0.04) | 0.06 (0.04) | 0.09 (0.04) | 0.01 (0.05) | |
TDR | |||||
GIN | AdaDetect | 0.10 (0.20) | 0.04 (0.10) | 0.42 (0.42) | 0.00 (0.00) |
CAD | 0.04 (0.12) | 0.03 (0.09) | 0.87 (0.20) | 0.02 (0.03) | |
DiffPool | AdaDetect | 0.04 (0.12) | 0.03 (0.06) | 0.62 (0.42) | 0.00 (0.00) |
CAD | 0.00 (0.01) | 0.01 (0.04) | 0.17 (0.24) | 0.00 (0.00) | |
DGCNN | AdaDetect | 0.27 (0.26) | 0.15 (0.12) | 0.95 (0.11) | 0.01 (0.02) |
CAD | 0.10 (0.17) | 0.05 (0.11) | 0.27 (0.26) | 0.00 (0.01) | |
WL | AdaDetect | 0.22 (0.18) | 0.12 (0.12) | 0.89 (0.07) | 0.01 (0.03) |
CAD | 0.29 (0.10) | 0.15 (0.10) | 0.49 (0.01) | 0.00 (0.00) | |