Beyond Ensemble Disagreement: Predictive STL Monitoring of RL Locomotion under Distribution Shift
Overview
This research investigates the reliability of ensemble disagreement as a safety indicator for reinforcement learning locomotion policies facing distribution shifts. Using frozen Ant-v5 and HalfCheetah-v5 policies, the authors evaluate predictive Signal Temporal Logic (STL) monitoring with closed-loop world-model rollouts across external-force, observation-noise, and mass shifts. Results show that while predictive robustness provides useful STL-specific ranking, it offers little additional value over current-state robustness metrics, which remain within 0.03 AUROC of predictive robustness under severe noise. The study cautions against treating ensemble disagreement as a reliable proxy for safety uncertainty, noting that violations persist at rates up to 64.2% under high noise conditions and that composite targets often lack positive violations in evaluated seed-condition slices.
Generated from attributed reports · Updated 56 minutes ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · robustness: 0.03 other · Basis not reported
“-state robustness without prediction comes within $0.03$ AUROC of predictive robustness. Penalizing predicted robustness by disagreement does not consistently improve ranking. Ant provides complementary per-specification evidence, but its composite target has no positive violations and its any-violation target is single-class in most evaluated seed-condition slices. Overall, predictive robustness provides useful STL-”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalBeyond Ensemble Disagreement: Predictive STL Monitoring of RL Locomotion under Distribution Shift
This research investigates the reliability of ensemble disagreement as a safety indicator for reinforcement learning locomotion policies facing distribution shifts. Using frozen Ant-v5 and HalfCheetah-v5 policies, the authors evaluate predictive Signal Temporal Logic (STL) monitoring with closed-loop world-model rollouts across external-force, observation-noise, and mass shifts. Results show that while predictive robustness provides useful STL-specific ranking, it offers little additional value over current-state robustness metrics, which remain within 0.03 AUROC of predictive robustness under severe noise. The study cautions against treating ensemble disagreement as a reliable proxy for safety uncertainty, noting that violations persist at rates up to 64.2% under high noise conditions and that composite targets often lack positive violations in evaluated seed-condition slices.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).