Matches in SemOpenAlex for { <https://semopenalex.org/work/W4376988891> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4376988891 abstract "AI systems empowered by reinforcement learning (RL) algorithms harbor the immense potential to catalyze societal advancement, yet their deployment is often impeded by significant safety concerns. Particularly in safety-critical applications, researchers have raised concerns about unintended harms or unsafe behaviors of unaligned RL agents. The philosophy of safe reinforcement learning (SafeRL) is to align RL agents with harmless intentions and safe behavioral patterns. In SafeRL, agents learn to develop optimal policies by receiving feedback from the environment, while also fulfilling the requirement of minimizing the risk of unintended harm or unsafe behavior. However, due to the intricate nature of SafeRL algorithm implementation, combining methodologies across various domains presents a formidable challenge. This had led to an absence of a cohesive and efficacious learning framework within the contemporary SafeRL research milieu. In this work, we introduce a foundational framework designed to expedite SafeRL research endeavors. Our comprehensive framework encompasses an array of algorithms spanning different RL domains and places heavy emphasis on safety elements. Our efforts are to make the SafeRL-related research process more streamlined and efficient, therefore facilitating further research in AI safety. Our project is released at: https://github.com/PKU-Alignment/omnisafe." @default.
- W4376988891 created "2023-05-18" @default.
- W4376988891 creator A5001682354 @default.
- W4376988891 creator A5010983432 @default.
- W4376988891 creator A5027669476 @default.
- W4376988891 creator A5037964892 @default.
- W4376988891 creator A5057799610 @default.
- W4376988891 creator A5062269449 @default.
- W4376988891 creator A5064143104 @default.
- W4376988891 creator A5072159736 @default.
- W4376988891 creator A5083150909 @default.
- W4376988891 creator A5090073634 @default.
- W4376988891 date "2023-05-16" @default.
- W4376988891 modified "2023-09-27" @default.
- W4376988891 title "OmniSafe: An Infrastructure for Accelerating Safe Reinforcement Learning Research" @default.
- W4376988891 doi "https://doi.org/10.48550/arxiv.2305.09304" @default.
- W4376988891 hasPublicationYear "2023" @default.
- W4376988891 type Work @default.
- W4376988891 citedByCount "0" @default.
- W4376988891 crossrefType "posted-content" @default.
- W4376988891 hasAuthorship W4376988891A5001682354 @default.
- W4376988891 hasAuthorship W4376988891A5010983432 @default.
- W4376988891 hasAuthorship W4376988891A5027669476 @default.
- W4376988891 hasAuthorship W4376988891A5037964892 @default.
- W4376988891 hasAuthorship W4376988891A5057799610 @default.
- W4376988891 hasAuthorship W4376988891A5062269449 @default.
- W4376988891 hasAuthorship W4376988891A5064143104 @default.
- W4376988891 hasAuthorship W4376988891A5072159736 @default.
- W4376988891 hasAuthorship W4376988891A5083150909 @default.
- W4376988891 hasAuthorship W4376988891A5090073634 @default.
- W4376988891 hasBestOaLocation W43769888911 @default.
- W4376988891 hasConcept C105339364 @default.
- W4376988891 hasConcept C111919701 @default.
- W4376988891 hasConcept C112930515 @default.
- W4376988891 hasConcept C115903868 @default.
- W4376988891 hasConcept C144133560 @default.
- W4376988891 hasConcept C154945302 @default.
- W4376988891 hasConcept C15744967 @default.
- W4376988891 hasConcept C17744445 @default.
- W4376988891 hasConcept C199539241 @default.
- W4376988891 hasConcept C2776889888 @default.
- W4376988891 hasConcept C2777363581 @default.
- W4376988891 hasConcept C41008148 @default.
- W4376988891 hasConcept C77805123 @default.
- W4376988891 hasConcept C97541855 @default.
- W4376988891 hasConcept C98045186 @default.
- W4376988891 hasConceptScore W4376988891C105339364 @default.
- W4376988891 hasConceptScore W4376988891C111919701 @default.
- W4376988891 hasConceptScore W4376988891C112930515 @default.
- W4376988891 hasConceptScore W4376988891C115903868 @default.
- W4376988891 hasConceptScore W4376988891C144133560 @default.
- W4376988891 hasConceptScore W4376988891C154945302 @default.
- W4376988891 hasConceptScore W4376988891C15744967 @default.
- W4376988891 hasConceptScore W4376988891C17744445 @default.
- W4376988891 hasConceptScore W4376988891C199539241 @default.
- W4376988891 hasConceptScore W4376988891C2776889888 @default.
- W4376988891 hasConceptScore W4376988891C2777363581 @default.
- W4376988891 hasConceptScore W4376988891C41008148 @default.
- W4376988891 hasConceptScore W4376988891C77805123 @default.
- W4376988891 hasConceptScore W4376988891C97541855 @default.
- W4376988891 hasConceptScore W4376988891C98045186 @default.
- W4376988891 hasLocation W43769888911 @default.
- W4376988891 hasOpenAccess W4376988891 @default.
- W4376988891 hasPrimaryLocation W43769888911 @default.
- W4376988891 hasRelatedWork W1562959674 @default.
- W4376988891 hasRelatedWork W1778072231 @default.
- W4376988891 hasRelatedWork W2096848550 @default.
- W4376988891 hasRelatedWork W2923653485 @default.
- W4376988891 hasRelatedWork W2937603438 @default.
- W4376988891 hasRelatedWork W2952472710 @default.
- W4376988891 hasRelatedWork W2957776456 @default.
- W4376988891 hasRelatedWork W4210912933 @default.
- W4376988891 hasRelatedWork W4221165949 @default.
- W4376988891 hasRelatedWork W4361026739 @default.
- W4376988891 isParatext "false" @default.
- W4376988891 isRetracted "false" @default.
- W4376988891 workType "article" @default.