On a 57-day, ten-router Internet2 traffic corpus, dense MLP forecaster TiDE cuts baseline prediction error by 30 to 42 percent relative to SARIMA and XGBoost, while anomaly masking gives only small robustness gains.
A signal analysis of network traffic anomalies,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing Anomaly Resilience in Research Networks: A Large-Scale Forecasting Benchmark for Dynamic Security Baselining
On a 57-day, ten-router Internet2 traffic corpus, dense MLP forecaster TiDE cuts baseline prediction error by 30 to 42 percent relative to SARIMA and XGBoost, while anomaly masking gives only small robustness gains.