Turdakov Denis Yur'evich

Publications in Math-Net.Ru

1. Convergence of a multilayer perceptron to histogram Bayesian regression
   
   Uspekhi Mat. Nauk, 80:6(486) (2025),  45–72
2. Method for training perceptron on tabular data with missing values
   
   Proceedings of ISP RAS, 37:6(2) (2025),  93–106
3. Iterative weak supervision with LLM-guided labeling function refinement
   
   Proceedings of ISP RAS, 37:6(2) (2025),  65–76
4. Developing a defence for large language models against adversarial attacks based on paraphrasing in a black-box scenario
   
   Proceedings of ISP RAS, 37:5 (2025),  195–204
5. Knowledge distillation in local-region for black-box adversarial examples
   
   Proceedings of ISP RAS, 37:4(2) (2025),  133–146
6. Extrapolation of the Bayesian classifier with an unknown support of the two-class mixture distribution
   
   Uspekhi Mat. Nauk, 79:6(480) (2024),  57–82
7. Development of a red-teaming dataset for defending large language models against attacks
   
   Proceedings of ISP RAS, 36:5 (2024),  143–152
8. Automating the process of responding to a tax request using weak supervision
   
   Proceedings of ISP RAS, 36:3 (2024),  203–212
9. Prospects for using a trusted information analytical system based on the Talisman platform using artificial intelligence methods to improve the efficiency of complex hardware systems
   
   Proceedings of ISP RAS, 36:3 (2024),  105–122
10. Detecting and eliminating covariate shifts in data for a more robust HDD failure prediction
    
    Zap. Nauchn. Sem. POMI, 540 (2024),  148–161
11. A study of graph neural networks for link prediction on vulnerability to membership attacks
    
    Zap. Nauchn. Sem. POMI, 530 (2023),  113–127
12. Trusted artificial intelligence: challenges and promising solutions
    
    Dokl. RAN. Math. Inf. Proc. Upr., 508 (2022),  13–18
13. Detecting influential users in social networks based on bipartite comments graph
    
    Proceedings of ISP RAS, 34:5 (2022),  127–142
14. Methods for determining the elements of the PQRST-complex of the electrocardiogram
    
    Proceedings of ISP RAS, 34:4 (2022),  229–240
15. Loss functions for train document image segmentation models
    
    Proceedings of ISP RAS, 34:2 (2022),  89–110
16. Assessment of the impact of non-architectural changes in the predictive model on the quality of ECG classification
    
    Proceedings of ISP RAS, 33:4 (2021),  87–98
17. Two step method for grouping news with similar topics
    
    Proceedings of ISP RAS, 32:4 (2020),  165–174
18. Diagnosis of left atrial and left ventricular hypertrophies using a deep neural network
    
    Proceedings of ISP RAS, 32:4 (2020),  141–154
19. Domain adaptation by proactive labeling
    
    Proceedings of ISP RAS, 31:5 (2019),  145–152
20. Methods for news items popularity estimation on early stages
    
    Proceedings of ISP RAS, 31:5 (2019),  137–144
21. Methods for information spread analysis
    
    Proceedings of ISP RAS, 30:6 (2018),  199–220
22. Active learning and crowdsourcing: a survey of annotation optimization methods
    
    Proceedings of ISP RAS, 30:2 (2018),  215–250
23. Approaches to estimate location of social network users based on social graph
    
    Proceedings of ISP RAS, 28:6 (2016),  185–196
24. Parallel modularity computation for directed weighted graphs with overlapping communities
    
    Proceedings of ISP RAS, 28:6 (2016),  153–170
25. Implementing Apache Spark jobs execution and Apache Spark cluster creation for OpenStack Sahara
    
    Proceedings of ISP RAS, 27:5 (2015),  35–48
26. Modern approaches to aspect-based sentiment analysis
    
    Proceedings of ISP RAS, 27:5 (2015),  5–22


27. Giant component of truncated scale-free graphs: theory and applications to creation of datasets
    
    Uspekhi Mat. Nauk, 80:6(486) (2025),  184–186