Matches in SemOpenAlex for { <https://semopenalex.org/work/W3003648996> ?p ?o ?g. }
- W3003648996 abstract "Reinforcement learning is a promising approach to synthesizing policies for robotics tasks. A key challenge is ensuring safety of the learned policy---e.g., that a walking robot does not fall over, or an autonomous car does not run into an obstacle. We focus on the setting where the dynamics are known, and the goal is to prove that a policy trained in simulation satisfies a given safety constraint. We build on an approach called shielding, which uses a backup policy to override the learned policy as needed to ensure safety. Our algorithm, called model predictive shielding (MPS), computes whether it is safe to use the learned policy on-the-fly instead of ahead-of-time. By doing so, our approach is computationally efficient, and can furthermore be used to ensure safety even in novel environments. Finally, we empirically demonstrate the benefits of our approach." @default.
- W3003648996 created "2020-02-07" @default.
- W3003648996 creator A5029243071 @default.
- W3003648996 date "2019-05-25" @default.
- W3003648996 modified "2023-09-27" @default.
- W3003648996 title "Safe Planning via Model Predictive Shielding" @default.
- W3003648996 cites W1577785134 @default.
- W3003648996 cites W1845972764 @default.
- W3003648996 cites W1945123189 @default.
- W3003648996 cites W1963790880 @default.
- W3003648996 cites W1993210515 @default.
- W3003648996 cites W2053572490 @default.
- W3003648996 cites W2104733512 @default.
- W3003648996 cites W2133404041 @default.
- W3003648996 cites W2139137304 @default.
- W3003648996 cites W2164479831 @default.
- W3003648996 cites W2296356821 @default.
- W3003648996 cites W2462906003 @default.
- W3003648996 cites W2616964725 @default.
- W3003648996 cites W2796284132 @default.
- W3003648996 cites W2803543472 @default.
- W3003648996 cites W2885163910 @default.
- W3003648996 cites W2889696653 @default.
- W3003648996 cites W2892521964 @default.
- W3003648996 cites W2952720101 @default.
- W3003648996 cites W2952905979 @default.
- W3003648996 cites W2962232176 @default.
- W3003648996 cites W2963575966 @default.
- W3003648996 cites W2964130946 @default.
- W3003648996 cites W2964161785 @default.
- W3003648996 cites W2973076431 @default.
- W3003648996 cites W3037935411 @default.
- W3003648996 cites W3120459386 @default.
- W3003648996 hasPublicationYear "2019" @default.
- W3003648996 type Work @default.
- W3003648996 sameAs 3003648996 @default.
- W3003648996 citedByCount "6" @default.
- W3003648996 countsByYear W30036489962019 @default.
- W3003648996 countsByYear W30036489962020 @default.
- W3003648996 countsByYear W30036489962022 @default.
- W3003648996 crossrefType "posted-content" @default.
- W3003648996 hasAuthorship W3003648996A5029243071 @default.
- W3003648996 hasConcept C112930515 @default.
- W3003648996 hasConcept C119599485 @default.
- W3003648996 hasConcept C119857082 @default.
- W3003648996 hasConcept C127413603 @default.
- W3003648996 hasConcept C144133560 @default.
- W3003648996 hasConcept C154945302 @default.
- W3003648996 hasConcept C172205157 @default.
- W3003648996 hasConcept C17744445 @default.
- W3003648996 hasConcept C199539241 @default.
- W3003648996 hasConcept C2265751 @default.
- W3003648996 hasConcept C26517878 @default.
- W3003648996 hasConcept C2775924081 @default.
- W3003648996 hasConcept C2776036281 @default.
- W3003648996 hasConcept C2776650193 @default.
- W3003648996 hasConcept C2780945871 @default.
- W3003648996 hasConcept C34413123 @default.
- W3003648996 hasConcept C38652104 @default.
- W3003648996 hasConcept C41008148 @default.
- W3003648996 hasConcept C77088390 @default.
- W3003648996 hasConcept C78519656 @default.
- W3003648996 hasConcept C90509273 @default.
- W3003648996 hasConcept C97541855 @default.
- W3003648996 hasConceptScore W3003648996C112930515 @default.
- W3003648996 hasConceptScore W3003648996C119599485 @default.
- W3003648996 hasConceptScore W3003648996C119857082 @default.
- W3003648996 hasConceptScore W3003648996C127413603 @default.
- W3003648996 hasConceptScore W3003648996C144133560 @default.
- W3003648996 hasConceptScore W3003648996C154945302 @default.
- W3003648996 hasConceptScore W3003648996C172205157 @default.
- W3003648996 hasConceptScore W3003648996C17744445 @default.
- W3003648996 hasConceptScore W3003648996C199539241 @default.
- W3003648996 hasConceptScore W3003648996C2265751 @default.
- W3003648996 hasConceptScore W3003648996C26517878 @default.
- W3003648996 hasConceptScore W3003648996C2775924081 @default.
- W3003648996 hasConceptScore W3003648996C2776036281 @default.
- W3003648996 hasConceptScore W3003648996C2776650193 @default.
- W3003648996 hasConceptScore W3003648996C2780945871 @default.
- W3003648996 hasConceptScore W3003648996C34413123 @default.
- W3003648996 hasConceptScore W3003648996C38652104 @default.
- W3003648996 hasConceptScore W3003648996C41008148 @default.
- W3003648996 hasConceptScore W3003648996C77088390 @default.
- W3003648996 hasConceptScore W3003648996C78519656 @default.
- W3003648996 hasConceptScore W3003648996C90509273 @default.
- W3003648996 hasConceptScore W3003648996C97541855 @default.
- W3003648996 hasLocation W30036489961 @default.
- W3003648996 hasOpenAccess W3003648996 @default.
- W3003648996 hasPrimaryLocation W30036489961 @default.
- W3003648996 hasRelatedWork W1528801453 @default.
- W3003648996 hasRelatedWork W2024216853 @default.
- W3003648996 hasRelatedWork W2053572490 @default.
- W3003648996 hasRelatedWork W2515079269 @default.
- W3003648996 hasRelatedWork W2761836782 @default.
- W3003648996 hasRelatedWork W2784755783 @default.
- W3003648996 hasRelatedWork W2921596426 @default.
- W3003648996 hasRelatedWork W2970602184 @default.
- W3003648996 hasRelatedWork W3044205027 @default.
- W3003648996 hasRelatedWork W3089438647 @default.
- W3003648996 hasRelatedWork W3092253435 @default.