Table 4 Coefficients set in the experiment
From: Dynamic access control method for SDP-based network environments
Variable | Weight | Definition |
|---|---|---|
\(\alpha\) | 0.30 | EMA smoothing coefficient of rewards and penalties |
\(\beta\) | 0.30 | Correction factor of reward values for past approach results |
\(\lambda\) | 0.40 | Correction factor for penalty points for past approach results |
\(\delta\) | 0.50 | Weight for PRV, RRI |
\(\varepsilon\) | 0.05–0.1 | Weight for PPV, RPI |
\(\zeta\) | 0.30 | Correction factor for the device compensation history SRV |
\(\eta\) | 0.30 | Correction factor for the device compensation history SPV |