Zobrazeno 1 - 10
of 11
pro vyhledávání: '"Panagiotis Kouvaros"'
Publikováno v:
AAAI
We introduce an efficient method for the verification of ReLU-based feed-forward neural networks. We derive an automated procedure that exploits dependency relations between the ReLU nodes, thereby pruning the search tree that needs to be considered
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d0780084990a3455466833ec2deb3452
https://doi.org/10.1609/aaai.v34i04.5729
https://doi.org/10.1609/aaai.v34i04.5729
Autor:
Denis Osipychev, Panagiotis Kouvaros, Trent Kyono, Dragos D. Margineantu, Alessio Lomuscio, Yang Zheng, Francesco Leofante
Publikováno v:
International Symposium on Formal Methods
Formal Methods ISBN: 9783030908690
FM
Formal Methods ISBN: 9783030908690
FM
Neural networks are being increasingly used for efficient decision making in the aircraft domain. Given the safety-critical nature of the applications involved, stringent safety requirements must be met by these networks. In this work we present a fo
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::db9dfbf4087d2488e0ac0f23fdb47515
http://hdl.handle.net/10044/1/94034
http://hdl.handle.net/10044/1/94034
Publikováno v:
AAMAS
We introduce a model for agent-environment systems where the agents are implemented via feed-forward ReLU neural networks and the environment is non-deterministic. We study the verification problem of such systems against CTL properties. We show that
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b6b492cd471e89f220627d9a1545a029
http://hdl.handle.net/10044/1/95834
http://hdl.handle.net/10044/1/95834
Autor:
Alessio Lomuscio, Panagiotis Kouvaros
Publikováno v:
IJCAI
We introduce an efficient method for the complete verification of ReLU-based feed-forward neural networks. The method implements branching on the ReLU states on the basis of a notion of dependency between the nodes. This results in dividing the origi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::21c97d9e06598bbe8c37ff6742a03c40
https://doi.org/10.24963/ijcai.2021/364
https://doi.org/10.24963/ijcai.2021/364
Publikováno v:
International Joint Conference on Artificial Intelligence (IJAC 2021)
IJCAI
IJCAI
We introduce an efficient and tight layer-based semidefinite relaxation for verifying local robustness of neural networks. The improved tightness is the result of the combination between semidefinite relaxations and linear cuts. We obtain a computati
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::53e6ca5403d279a185c49ca4b8ae1535
http://hdl.handle.net/10044/1/89706
http://hdl.handle.net/10044/1/89706
Publikováno v:
Software Engineering and Formal Methods ISBN: 9783030921231
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::561bae85016183049f32b23611afb3a0
https://doi.org/10.1007/978-3-030-92124-8_26
https://doi.org/10.1007/978-3-030-92124-8_26
Publikováno v:
KR
We investigate the problem of verifying the strategic properties of multi-agent systems equipped with machine learning-based perception units. We introduce a novel model of agents comprising both a perception system implemented via feed-forward neura
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d5a5976a69c62f3226fb89378e6d29df
https://doi.org/10.24963/kr.2020/3
https://doi.org/10.24963/kr.2020/3
Publikováno v:
IJCAI
We study the problem of determining the robustness of a multi-agent system of unbounded size against specifications expressed in a temporal-epistemic logic. We introduce a procedure to synthesise automatically the maximal ratio of faulty agents that
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fe54c01a2c900e3007b0ae700f2a7b7b
https://doi.org/10.24963/ijcai.2018/45
https://doi.org/10.24963/ijcai.2018/45
Publikováno v:
26th International Joint Conference on Artificial Intelligence (IJCAI 2017)
26th International Joint Conference on Artificial Intelligence (IJCAI 2017), Aug 2017, Melbourne, Australia. pp.98--104
Scopus-Elsevier
IJCAI
26th International Joint Conference on Artificial Intelligence (IJCAI 2017), Aug 2017, Melbourne, Australia. pp.98--104
Scopus-Elsevier
IJCAI
We introduce parameterised data-aware multi-agent systems, a formalism to reason about the temporal-epistemic properties of arbitrarily large collections of homogeneous agents, each operating on an infinite data domain. We show that their parameteris
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2d9527a25cffc0a095be4dcb9238d588
https://hal.science/hal-01629367
https://hal.science/hal-01629367
Autor:
Panagiotis Kouvaros, Alessio Lomuscio
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 31
We define a class of parameterised infinite state multi-agent systems (MAS) that is unbounded in both the number of agents composing the system and the domain of the variables encoding the agents. We analyse their verification problem by combining an
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e4e8581cde3ef3b4213497082ddb1380
https://doi.org/10.1609/aaai.v31i1.10710
https://doi.org/10.1609/aaai.v31i1.10710