On Cryptographic Attacks Using Backdoors for SAT

Authors: Alexander Semenov, Oleg Zaikin, Ilya Otpuschennikov, Stepan Kochemazov, Alexey Ignatiev

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on weakened variants of the renowned encryption algorithms exhibit advantage of the proposed approach compared to the state of the art in terms of the estimated hardness of the resulting guess-and-determine attacks.
Researcher Affiliation Academia 1 Matrosov Institute for System Dynamics and Control Theory SB RAS, Irkutsk, Russia 2 LASIGE, Faculdade de Ciˆencias, Universidade de Lisboa, Portugal
Pseudocode Yes Algorithm 1 shows the pseudo-code of the algorithm.
Open Source Code No The paper does not include an unambiguous statement that the authors are releasing the source code for the work described, nor does it provide a direct link to a code repository.
Open Datasets No The paper uses cryptographic ciphers (Trivium, AES-128, Magma) as benchmarks and mentions using 'known plaintext' for cryptanalysis problems, but it does not specify or provide access information for a publicly available or open dataset in the conventional sense. These are not datasets for training/testing ML models, but rather systems for cryptanalysis.
Dataset Splits No The paper discusses the use of 'random samples' for Monte-Carlo estimation of probability PB(t) and 'r' observed outputs for the G-a-D attack, which are not standard training/validation/test dataset splits typically used in machine learning for model reproduction.
Hardware Specification Yes All experiments were run on 10 nodes of a computing cluster, each node being equipped with two Intel Xeon E52695 v4 CPUs and 128 GB RAM.
Software Dependencies No The paper mentions using the 'ROKK SAT solver' and states the authors/year (Yasumoto and Okuwaga 2014) but does not provide a specific version number for the software.
Experiment Setup Yes During the resistance function minimization, we used random samples of size 1000. When searching for best IBSs, the values of PB and t typically were in intervals PB [0.05, 1], t [1s, 200s].