2 papers
cs.LG2026
Robustness Meets Uncertainty: Evidential Adversarial Training for Robust Selective Classification
Nicolas Sournac, Ahmed Baha Ben Jmaa, Bertrand Braeckeveldt
Safety-critical applications require classifiers that are both robust and reliable. Adversarial training is a widely adopted defense for improving robustness in deep neural network…
physics.optics2025
Timekeeping precision enhancements at constant power
K. J. H. Peters, B. Braeckeveldt, B. Maes +1
We demonstrate precision enhancements in a timekeeping device at constant power and regardless of the operation frequency. Our timekeeping device is a laser-driven coupled-cavity s…