Matches in SemOpenAlex for { <https://semopenalex.org/work/W3188777742> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3188777742 abstract "Multi-agent control problems constitute an interesting area of application for deep reinforcement learning models with continuous action spaces. Such real-world applications, however, typically come with critical safety constraints that must not be violated. In order to ensure safety, we enhance the well-known multi-agent deep deterministic policy gradient (MADDPG) framework by adding a safety layer to the deep policy network. In particular, we extend the idea of linearizing the single-step transition dynamics, as was done for single-agent systems in Safe DDPG (Dalal et al., 2018), to multi-agent settings. We additionally propose to circumvent infeasibility problems in the action correction step using soft constraints (Kerrigan & Maciejowski, 2000). Results from the theory of exact penalty functions can be used to guarantee constraint satisfaction of the soft constraints under mild assumptions. We empirically find that the soft formulation achieves a dramatic decrease in constraint violations, making safety available even during the learning procedure." @default.
- W3188777742 created "2021-08-16" @default.
- W3188777742 creator A5016531134 @default.
- W3188777742 creator A5022640490 @default.
- W3188777742 creator A5031540264 @default.
- W3188777742 creator A5039882020 @default.
- W3188777742 creator A5043239271 @default.
- W3188777742 creator A5089301599 @default.
- W3188777742 date "2021-08-09" @default.
- W3188777742 modified "2023-09-23" @default.
- W3188777742 title "Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces." @default.
- W3188777742 cites W1572804421 @default.
- W3188777742 cites W1757796397 @default.
- W3188777742 cites W2027579135 @default.
- W3188777742 cites W2145339207 @default.
- W3188777742 cites W2159307101 @default.
- W3188777742 cites W2173248099 @default.
- W3188777742 cites W2555811267 @default.
- W3188777742 cites W2575705757 @default.
- W3188777742 cites W2623431351 @default.
- W3188777742 cites W2784465508 @default.
- W3188777742 cites W2913300629 @default.
- W3188777742 cites W2959221924 @default.
- W3188777742 cites W2962775887 @default.
- W3188777742 cites W2963575966 @default.
- W3188777742 cites W2963881016 @default.
- W3188777742 cites W2964121744 @default.
- W3188777742 cites W2982041656 @default.
- W3188777742 cites W3095315965 @default.
- W3188777742 cites W3100944043 @default.
- W3188777742 hasPublicationYear "2021" @default.
- W3188777742 type Work @default.
- W3188777742 sameAs 3188777742 @default.
- W3188777742 citedByCount "0" @default.
- W3188777742 crossrefType "posted-content" @default.
- W3188777742 hasAuthorship W3188777742A5016531134 @default.
- W3188777742 hasAuthorship W3188777742A5022640490 @default.
- W3188777742 hasAuthorship W3188777742A5031540264 @default.
- W3188777742 hasAuthorship W3188777742A5039882020 @default.
- W3188777742 hasAuthorship W3188777742A5043239271 @default.
- W3188777742 hasAuthorship W3188777742A5089301599 @default.
- W3188777742 hasConcept C121332964 @default.
- W3188777742 hasConcept C126255220 @default.
- W3188777742 hasConcept C154945302 @default.
- W3188777742 hasConcept C199622910 @default.
- W3188777742 hasConcept C2524010 @default.
- W3188777742 hasConcept C2776036281 @default.
- W3188777742 hasConcept C2780791683 @default.
- W3188777742 hasConcept C33923547 @default.
- W3188777742 hasConcept C41008148 @default.
- W3188777742 hasConcept C44616089 @default.
- W3188777742 hasConcept C49937458 @default.
- W3188777742 hasConcept C62520636 @default.
- W3188777742 hasConcept C97541855 @default.
- W3188777742 hasConceptScore W3188777742C121332964 @default.
- W3188777742 hasConceptScore W3188777742C126255220 @default.
- W3188777742 hasConceptScore W3188777742C154945302 @default.
- W3188777742 hasConceptScore W3188777742C199622910 @default.
- W3188777742 hasConceptScore W3188777742C2524010 @default.
- W3188777742 hasConceptScore W3188777742C2776036281 @default.
- W3188777742 hasConceptScore W3188777742C2780791683 @default.
- W3188777742 hasConceptScore W3188777742C33923547 @default.
- W3188777742 hasConceptScore W3188777742C41008148 @default.
- W3188777742 hasConceptScore W3188777742C44616089 @default.
- W3188777742 hasConceptScore W3188777742C49937458 @default.
- W3188777742 hasConceptScore W3188777742C62520636 @default.
- W3188777742 hasConceptScore W3188777742C97541855 @default.
- W3188777742 hasOpenAccess W3188777742 @default.
- W3188777742 hasRelatedWork W1553983339 @default.
- W3188777742 hasRelatedWork W1598210495 @default.
- W3188777742 hasRelatedWork W2063515892 @default.
- W3188777742 hasRelatedWork W2107166738 @default.
- W3188777742 hasRelatedWork W2109992583 @default.
- W3188777742 hasRelatedWork W2126242014 @default.
- W3188777742 hasRelatedWork W2181620480 @default.
- W3188777742 hasRelatedWork W2345155550 @default.
- W3188777742 hasRelatedWork W2750073679 @default.
- W3188777742 hasRelatedWork W2900892283 @default.
- W3188777742 hasRelatedWork W292588544 @default.
- W3188777742 hasRelatedWork W2964046206 @default.
- W3188777742 hasRelatedWork W3002224236 @default.
- W3188777742 hasRelatedWork W3012839911 @default.
- W3188777742 hasRelatedWork W3091364228 @default.
- W3188777742 hasRelatedWork W3122941067 @default.
- W3188777742 hasRelatedWork W3126230687 @default.
- W3188777742 hasRelatedWork W3132081946 @default.
- W3188777742 hasRelatedWork W2238370166 @default.
- W3188777742 hasRelatedWork W2734655874 @default.
- W3188777742 isParatext "false" @default.
- W3188777742 isRetracted "false" @default.
- W3188777742 magId "3188777742" @default.
- W3188777742 workType "article" @default.