Search papers, labs, and topics across Lattice.
Fraunhofer Institute for Cognitive Systems IKS Garching
2
0
3
CUBICS reveals that context-aware performance estimation can significantly improve the reliability of safety-critical ML components by avoiding oversimplified failure models.
Real-time safety assurance gets a boost: this method dynamically adjusts confidence in safety arguments based on runtime performance, reacting swiftly to violations.
