AI Paper Atlas

Research map / Reinforcement learning / Policy gradient and actor-critic

Probabilistic Safety Verification of Neural Policies via Predicate Abstraction

Marcel Vinzent, Holger Hermanns, Jörg Hoffmann

AAAI 2026 · Reinforcement learning › Policy gradient and actor-critic · approach: Safe policy improvement

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use prob-CEGAR-PPA-inc to derive boundsneural policies · 32 benchmark instancescompletion: complete on 16 of 32Mixed
handle probabilistic transitions in abstract state space computation instead of non-probabilistic abstractionPPA · abstract state space computation timetimeNo clear effect
use prob-CEGAR-PPA-pu instead of incremental search for a fixed puprob-CEGAR-PPA · 4 Blocks CA NN32 pu=0.5verification time: 1145 -> 399 sImproved
use incremental search for pu close to true unsafety probabilityprob-CEGAR-PPA · 4 Blocks CA NN32 app, pu=0.5feasibility: timeout -> 2056 sImproved
add applicability filtering to policyprob-CEGAR-PPA · counterexample analysistimeouts: 9 timeouts out of 13Hurt
use prob-CEGAR-PPA-inc instead of probabilistic model checkerssafety verification · benchmarkssolved instances: STORM explicit: 180 s vs under 80 sImproved
vary step size p_delta in {0.01,0.1} vs 0.05 (ablation)prob-CEGAR-PPA-inc · runtimeruntimeNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy