Disentangled Dynamic Deviation Transformer Networks for Multivariate Time Series Anomaly Detection

. 2023 Jan 18;23(3):1104. doi: 10.3390/s23031104

Algorithm 1 POT-MoM

Input:
input data $\{l_{1}, l_{2}, \dots, l_{w}\}$ , risk q
Output:
initial threshold $τ$ , final threshold $τ_{f i n a l}$
1:
$τ \leftarrow S e t I n i t i a l T h r e s h o l d (\{l_{1}, l_{2}, \dots, l_{w}\})$
2:
$Y_{t} \leftarrow \{L_{i} - τ | L_{i} > τ\}$
3:
$\hat{α}, \hat{β} \leftarrow M o M (Y_{t})$
4:
$τ_{f i n a l} \leftarrow C a l c F i n a l T h r e d h o l d (q, \hat{α}, \hat{β}, n, N_{t}, t)$