4 papers
Learning to Navigate Under Imperfect Perception: Conformalised Segmentation for Safe Reinforcement Learning
Daniel Bethell, Simos Gerasimou, Radu Calinescu +1
Reliable navigation in safety-critical environments requires both accurate hazard perception and principled uncertainty handling to strengthen downstream safety handling. Despite t…
Safe But Not Sorry: Reducing Over-Conservatism in Safety Critics via Uncertainty-Aware Modulation
Daniel Bethell, Simos Gerasimou, Radu Calinescu +1
Ensuring the safe exploration of reinforcement learning (RL) agents is critical for deployment in real-world systems. Yet existing approaches struggle to strike the right balance:…
Guided Uncertainty Learning Using a Post-Hoc Evidential Meta-Model
Charmaine Barker, Daniel Bethell, Simos Gerasimou
Reliable uncertainty quantification remains a major obstacle to the deployment of deep learning models under distributional shift. Existing post-hoc approaches that retrofit pretra…
Robust Adversarial Quantification via Conflict-Aware Evidential Deep Learning
Charmaine Barker, Daniel Bethell, Simos Gerasimou
Reliability of deep learning models is critical for deployment in high-stakes applications, where out-of-distribution or adversarial inputs may lead to detrimental outcomes. Eviden…