Researchers have introduced MuViS-C, a novel benchmark designed to evaluate the robustness of learning-based virtual sensing systems when faced with sensor failures. The benchmark covers ten distinct sensor failure modes across six domains and nine datasets, assessing performance degradation and worst-case fragility. Experiments revealed that all tested architectures, including those based on attention mechanisms, significantly degrade under corruption, with gradient-boosted trees showing strong robustness. The study also found that dedicated robustification strategies can improve performance but may negatively impact nominal accuracy. AI
IMPACT This benchmark will help researchers develop more reliable AI systems for cyber-physical applications by highlighting vulnerabilities to sensor failures.
RANK_REASON The cluster describes a new benchmark and research paper published on arXiv.
Read on Hugging Face Daily Papers →
- arXiv
- attention
- CatalyzeX
- convolutional neural network
- DagsHub
- Gotit.pub
- Gradient Boosted Trees
- Hugging Face
- MLP-mixing
- MuViS-C
- ScienceCast
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
