FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3039444993> ?p ?o ?g. }

• W3039444993 abstract "Reinforcement learning (RL) has achieved tremendous progress in solving various sequential decision-making problems, e.g., control tasks in robotics. However, RL methods often fail to generalize to safety-critical scenarios since policies are overfitted to training environments. Previously, robust adversarial reinforcement learning (RARL) was proposed to train an adversarial network that applies disturbances to a system, which improves robustness in test scenarios. A drawback of neural-network-based adversaries is that integrating system requirements without handcrafting sophisticated reward signals is difficult. Safety falsification methods allow one to find a set of initial conditions as well as an input sequence, such that the system violates a given property formulated in temporal logic. In this paper, we propose falsification-based RARL (FRARL), the first generic framework for integrating temporal-logic falsification in adversarial learning to improve policy robustness. With falsification method, we do not need to construct an extra reward function for the adversary. We evaluate our approach on a braking assistance system and an adaptive cruise control system of autonomous vehicles. Experiments show that policies trained with a falsification-based adversary generalize better and show less violation of the safety specification in test scenarios than the ones trained without an adversary or with an adversarial network." @default.
• W3039444993 created "2020-07-10" @default.
• W3039444993 creator A5005383495 @default.
• W3039444993 creator A5044430490 @default.
• W3039444993 creator A5058010200 @default.
• W3039444993 date "2020-07-01" @default.
• W3039444993 modified "2023-10-16" @default.
• W3039444993 title "Falsification-Based Robust Adversarial Reinforcement Learning" @default.
• W3039444993 cites W1956142090 @default.
• W3039444993 cites W1965455100 @default.
• W3039444993 cites W1965555277 @default.
• W3039444993 cites W1966224165 @default.
• W3039444993 cites W198091129 @default.
• W3039444993 cites W2026629052 @default.
• W3039444993 cites W2031397756 @default.
• W3039444993 cites W2039287452 @default.
• W3039444993 cites W2041422323 @default.
• W3039444993 cites W2049399166 @default.
• W3039444993 cites W2105078254 @default.
• W3039444993 cites W2112656403 @default.
• W3039444993 cites W2124267516 @default.
• W3039444993 cites W2131399618 @default.
• W3039444993 cites W2145339207 @default.
• W3039444993 cites W2172184261 @default.
• W3039444993 cites W2257979135 @default.
• W3039444993 cites W2334461743 @default.
• W3039444993 cites W2396555304 @default.
• W3039444993 cites W2462906003 @default.
• W3039444993 cites W2561466143 @default.
• W3039444993 cites W2580909119 @default.
• W3039444993 cites W2586751528 @default.
• W3039444993 cites W2602963933 @default.
• W3039444993 cites W2604394466 @default.
• W3039444993 cites W2736601468 @default.
• W3039444993 cites W2740508249 @default.
• W3039444993 cites W2749567864 @default.
• W3039444993 cites W2773691349 @default.
• W3039444993 cites W2896642734 @default.
• W3039444993 cites W2962927323 @default.
• W3039444993 cites W2963184621 @default.
• W3039444993 cites W2963207607 @default.
• W3039444993 cites W2963373786 @default.
• W3039444993 cites W2964108292 @default.
• W3039444993 cites W2964121744 @default.
• W3039444993 cites W2964173023 @default.
• W3039444993 cites W2967292964 @default.
• W3039444993 cites W2968831808 @default.
• W3039444993 cites W2968983352 @default.
• W3039444993 cites W2991627542 @default.
• W3039444993 cites W3098659183 @default.
• W3039444993 cites W3100789280 @default.
• W3039444993 cites W3163244177 @default.
• W3039444993 cites W618136976 @default.
• W3039444993 cites W93764711 @default.
• W3039444993 doi "https://doi.org/10.48550/arxiv.2007.00691" @default.
• W3039444993 hasPublicationYear "2020" @default.
• W3039444993 type Work @default.
• W3039444993 sameAs 3039444993 @default.
• W3039444993 citedByCount "1" @default.
• W3039444993 countsByYear W30394449932020 @default.
• W3039444993 crossrefType "posted-content" @default.
• W3039444993 hasAuthorship W3039444993A5005383495 @default.
• W3039444993 hasAuthorship W3039444993A5044430490 @default.
• W3039444993 hasAuthorship W3039444993A5058010200 @default.
• W3039444993 hasBestOaLocation W30394449931 @default.
• W3039444993 hasConcept C104317684 @default.
• W3039444993 hasConcept C119857082 @default.
• W3039444993 hasConcept C14036430 @default.
• W3039444993 hasConcept C154945302 @default.
• W3039444993 hasConcept C177264268 @default.
• W3039444993 hasConcept C185592680 @default.
• W3039444993 hasConcept C199360897 @default.
• W3039444993 hasConcept C37736160 @default.
• W3039444993 hasConcept C38652104 @default.
• W3039444993 hasConcept C41008148 @default.
• W3039444993 hasConcept C41065033 @default.
• W3039444993 hasConcept C50644808 @default.
• W3039444993 hasConcept C55493867 @default.
• W3039444993 hasConcept C63479239 @default.
• W3039444993 hasConcept C78458016 @default.
• W3039444993 hasConcept C86803240 @default.
• W3039444993 hasConcept C97541855 @default.
• W3039444993 hasConceptScore W3039444993C104317684 @default.
• W3039444993 hasConceptScore W3039444993C119857082 @default.
• W3039444993 hasConceptScore W3039444993C14036430 @default.
• W3039444993 hasConceptScore W3039444993C154945302 @default.
• W3039444993 hasConceptScore W3039444993C177264268 @default.
• W3039444993 hasConceptScore W3039444993C185592680 @default.
• W3039444993 hasConceptScore W3039444993C199360897 @default.
• W3039444993 hasConceptScore W3039444993C37736160 @default.
• W3039444993 hasConceptScore W3039444993C38652104 @default.
• W3039444993 hasConceptScore W3039444993C41008148 @default.
• W3039444993 hasConceptScore W3039444993C41065033 @default.
• W3039444993 hasConceptScore W3039444993C50644808 @default.
• W3039444993 hasConceptScore W3039444993C55493867 @default.
• W3039444993 hasConceptScore W3039444993C63479239 @default.
• W3039444993 hasConceptScore W3039444993C78458016 @default.
• W3039444993 hasConceptScore W3039444993C86803240 @default.
• W3039444993 hasConceptScore W3039444993C97541855 @default.
• W3039444993 hasLocation W30394449931 @default.

• next