Zobrazeno 1 - 10
of 881
pro vyhledávání: '"Langé, Martin"'
We analyse the complexity of the satisfiability problem (SAT) for transformer encoders (TE), naturally occurring in formal verification or interpretation tasks. We find that SAT is undecidable when considering TE as they are commonly studied in the e
Externí odkaz:
http://arxiv.org/abs/2405.18548
Autor:
Laux, Hans Dieter
Publikováno v:
Erdkunde, 2016 Oct 01. 70(4), 371-372.
Externí odkaz:
https://www.jstor.org/stable/44280256
Verifying properties and interpreting the behaviour of deep neural networks (DNN) is an important task given their ubiquitous use in applications, including safety-critical ones, and their black-box nature. We propose an automata-theoric approach to
Externí odkaz:
http://arxiv.org/abs/2211.01022
Publikováno v:
EPTCS 370, 2022, pp. 17-33
Otto's Theorem characterises the bisimulation-invariant PTIME queries over graphs as exactly those that can be formulated in the polyadic mu-calculus, hinging on the Immerman-Vardi Theorem which characterises PTIME (over ordered structures) by First-
Externí odkaz:
http://arxiv.org/abs/2209.10311
Autor:
Sälzer, Marco, Lange, Martin
Output reachability and adversarial robustness are among the most relevant safety properties of neural networks. We show that in the context of Message Passing Neural Networks (MPNN), a common Graph Neural Network (GNN) model, formal verification is
Externí odkaz:
http://arxiv.org/abs/2206.05070
Autor:
Sälzer, Marco, Lange, Martin
Publikováno v:
Fundamenta Informaticae, Volume 189, Issues 3-4: Reachability Problems 2020 and 2021 (October 14, 2023) fi:9219
We investigate the complexity of the reachability problem for (deep) neural networks: does it compute valid output given some valid input? It was recently claimed that the problem is NP-complete for general neural networks and specifications over the
Externí odkaz:
http://arxiv.org/abs/2203.07941
Autor:
Sälzer, Marco, Lange, Martin
We investigate the complexity of the reachability problem for (deep) neural networks: does it compute valid output given some valid input? It was recently claimed that the problem is NP-complete for general neural networks and conjunctive input/outpu
Externí odkaz:
http://arxiv.org/abs/2108.13179
Publikováno v:
EPTCS 339, 2021, pp. 10-26
We investigate the expressive power of the two main kinds of program logics for complex, non-regular program properties found in the literature: those extending propositional dynamic logic (PDL), and those extending the modal mu-calculus. This is ins
Externí odkaz:
http://arxiv.org/abs/2108.10490
Autor:
Graham, Keith, Lienau, Philip, Bader, Benjamin, Prechtl, Stefan, Naujoks, Jan, Lesche, Ralf, Weiske, Joerg, Kuehnlenz, Julia, Brzezinka, Krzysztof, Potze, Lisette, Zanconato, Francesca, Nicke, Barbara, Montebaur, Anna, Bone, Wilhelm, Golfier, Sven, Kaulfuss, Stefan, Kopitz, Charlotte, Pilari, Sabine, Steuber, Holger, Hayat, Sikander, Kamburov, Atanas, Steffen, Andreas, Schlicker, Andreas, Buchgraber, Philipp, Braeuer, Nico, Font, Nuria Aiguabella, Heinrich, Tobias, Kuhnke, Lara, Nowak-Reppel, Katrin, Stresemann, Carlo, Steigemann, Patrick, Walter, Annette O., Blotta, Simona, Ocker, Matthias, Lakner, Ashley, von Nussbaum, Franz, Mumberg, Dominik, Eis, Knut, Piccolo, Stefano, Lange, Martin
Publikováno v:
In Cell Chemical Biology 18 July 2024 31(7):1247-1263
Autor:
Fiedler, Magdalena, Lange, Martin A., Hippauf, Felix, Dörfler, Susanne, Althues, Holger, Zeier, Wolfgang G., Kaskel, Stefan
Publikováno v:
In Carbon 30 June 2024 227