MILD: Multi-Intent Learning and Disambiguation for Proactive Failure Prediction in Intent-based Networking
This addresses proactive failure prediction and disambiguation for intent-based networking, offering incremental improvements in lead time and accuracy.
The paper tackles the problem of ambiguous root-cause identification in multi-intent networks where faults cause co-drift, by proposing MILD, a proactive framework that predicts failures and disambiguates intents, resulting in 3.8%–92.5% longer remediation lead times and 9.4%–45.8% higher disambiguation accuracy over baselines.
In multi-intent intent-based networks, a single fault can trigger co-drift where multiple intents exhibit symptomatic KPI degradation, creating ambiguity about the true root-cause intent. We present MILD, a proactive framework that reformulates intent assurance from reactive drift detection to fixed-horizon failure prediction with intent-level disambiguation. MILD uses a teacher-augmented Mixture-of-Experts where a gated disambiguation module identifies the root-cause intent while per-intent heads output calibrated risk scores. On a benchmark with non-linear failures and co-drifts, MILD provides 3.8\%--92.5\% longer remediation lead time and improves intent-level root-cause disambiguation accuracy by 9.4\%--45.8\% over baselines. MILD also provides per-alert KPI explanations, enabling actionable diagnosis.