A new training method uses scaled hybrid zonotopes to turn exact ReLU reachability into a differentiable loss, enabling verified avoidance of non-convex unsafe sets.
Challenges of real-world reinforcement learning: definitions, benchmarks and analysis,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Provably-Safe Neural Network Training Using Hybrid Zonotope Reachability Analysis
A new training method uses scaled hybrid zonotopes to turn exact ReLU reachability into a differentiable loss, enabling verified avoidance of non-convex unsafe sets.