Services:
Training
Development
System Administration
Security Audits
Contact:
Current Position
Department of Computer Science
University of Bucharest
Lecturer interested in signal processing, operating systems and security. Currently teaching "Operating Systems" and "Using Operating Systems" courses.
Past Positions
Department of Automatic Control and Computers
University Politehnica of Bucharest
Postdoc involved in two research projects. I also taught "Numerical Methods", "Parallel and distributed processing of data and knowledge", and "Modelling and Simulation".
Sparse representations in signal processing
IDEI 2011 project, January 2014 - September 2016
Researching dictionary learning for compressed sensing applications with a focus on parallel GPU implementations.
Set-theoretic approaches for fault tolerant control of complex systems
TE 2015 project, September 2015 - December 2017
Investigating sensor placement using sparse representations and dictionary learning with applications in water distribution networks.
Publications
Books:
[1] | B. Dumitrescu and P. Irofti, Dictionary Learning Algorithms and Applications, Springer, 2018. [ bib | DOI ] |
Chapters:
[1] | A. Pătrașcu, C. Păduraru, and P. Irofti, “Stochastic Proximal Gradient Algorithm with Minibatches. Application to Large Scale Learning Models,” in Enabling AI applications in Data Science, pp. 3--25. Springer, 2020. [ bib | DOI ] |
[2] | P. Irofti, A. Băltoiu, and A. Pătrașcu, “Fraud Detection in Networks,” in Enabling AI applications in Data Science, pp. 517--536. Springer, 2020. [ bib | DOI ] |
Journal:
[1] | P. Irofti, L. Romero-Ben, F. Stoican, and V. Puig, “Learning Dictionaries from Physical-Based Interpolation for Water Network Leak Localization,” IEEE Transactions on Control Systems Technology, pp. 1--12, 2023. [ bib | DOI ] |
[2] | A. Pătrașcu and P. Irofti, “On finite termination of an inexact Proximal Point algorithm,” Applied Mathematics Letters, vol. 134, pp. 108348, 2022. [ bib | DOI | http ] |
[3] | A. Pătrașcu and P. Irofti, “Stochastic proximal splitting algorithm for composite minimization,” Optimization Letters, pp. 1--19, 2021. [ bib | DOI | .pdf ] |
[4] | P. Irofti, F. Stoican, and V. Puig, “Fault Handling in Large Water Networks with Online Dictionary Learning,” Journal of Process Control, vol. 94, pp. 46--57, 2020. [ bib | DOI | .pdf ] |
[5] | F. Stoican and P. Irofti, “Aiding Dictionary Learning Through Multi-Parametric Sparse Representation,” Algorithms, vol. 12, no. 7, pp. 131, 2019. [ bib | .pdf ] |
[6] | B. Dumitrescu and P. Irofti, “Regularized K-SVD,” IEEE Signal Processing Letters, vol. 24, no. 3, pp. 309--313, March 2017. [ bib | DOI | .pdf ] |
[7] | P. Irofti, “Efficient Parallel Implementation for Single Block Orthogonal Dictionary Learning,” Journal of Control Engineering and Applied Informatics, vol. 18, no. 3, pp. 101--108, 2016. [ bib | .pdf ] |
[8] | P. Irofti, “Efficient Dictionary Learning Implementation on the GPU Using OpenCL,” U.P.B. Scientific Bulletin, Series C, vol. 78, no. 3, pp. 39--50, 2016. [ bib | .pdf ] |
Conference:
[1] | L. Romero-Ben, P. Irofti, F. Stoican, and V. Puig, “Nodal Hydraulic Head Estimation through Unscented Kalman Filter for Data-driven Leak Localization in Water Networks,” 12th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes SAFEPROCESS 2024 (IFAC-PapersOnLine), pp. 1--6, 2024. [ bib | http ] |
[2] | P. Irofti, “Pinky: A Modern Malware-oriented Dynamic Information Retrieval Tool,” in International Conference on Information Technology and Communications Security. Springer, 2023, pp. 65--78. [ bib | DOI | http ] |
[3] | A. Stancu, P. Irofti, and I. Leuștean, “OpenBSD formal driver verification with SeL4,” in International Conference on Information Technology and Communications Security. Springer, 2023, pp. 144--156. [ bib | DOI | http ] |
[4] | R. Bălucea and P. Irofti, “Software Mitigation of RISC-V Spectre Attacks,” in International Conference on Information Technology and Communications Security. Springer, 2023, pp. 51--64. [ bib | DOI | http ] |
[5] | P. Irofti, A. Pătrașcu, and A.I. Hîji, “Unsupervised Abnormal Traffic Detection through Topological Flow Analysis,” in 2022 14th International Conference on Communications (COMM). 2022, pp. 1--6, IEEE. [ bib | DOI | http ] |
[6] | P. Irofti, L. Romero-Ben, F. Stoican, and V. Puig, “Data-driven Leak Localization in Water Distribution Networks via Dictionary Learning and Graph-based Interpolation,” in CCTA 2022 - 2022 IEEE International Conference on Control Technology and Applications (CCTA). 2022, pp. 1265--1270, IEEE. [ bib | http ] |
[7] | P. Irofti, C. Rusu, and A. Pătrașcu, “Dictionary Learning with Uniform Sparse Representations for Anomaly Detection,” in ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 2022, pp. 3378--3382, IEEE. [ bib | DOI | http ] |
[8] | C. Rusu and P. Irofti, “Efficient and Parallel Separable Dictionary Learning,” in Proceedings of the IEEE 2021 27th International Conference on Parallel and Distributed Systems (ICPADS). 2021, pp. 1--6, IEEE Computer Society. [ bib | http ] |
[9] | A. Băltoiu, A. Pătrașcu, and P. Irofti, “Graph Anomaly Detection Using Dictionary Learning,” The 21st World Congress of the International Federation of Automatic Control (IFAC-PapersOnLine), vol. 53, no. 2, pp. 3551--3558, 2020. [ bib ] |
[10] | P. Irofti and A. Băltoiu, “Malware Identification with Dictionary Learning,” in 27th European Signal Processing Conference, 2019, pp. 1--5. [ bib | .pdf ] |
[11] | P. Irofti and B. Dumitrescu, “Pairwise Approximate K-SVD,” in Acoustics Speech and Signal Processing (ICASSP), 2019 IEEE International Conference on, 2019, pp. 3677--3681. [ bib | .pdf ] |
[12] | P. Irofti and F. Stoican, “Dictionary Learning Strategies for Sensor Placement and Leakage Isolation in Water Networks,” in The 20th World Congress of the International Federation of Automatic Control, 2017, pp. 1589--1594. [ bib | .pdf ] |
[13] | B. Dumitrescu and P. Irofti, “Low Dimensional Subspace Finding via Size-Reducing Dictionary Learning,” in 2016 IEEE International Workshop on Machine Learning for Signal Processing (MLSP), 2016, pp. 1--6. [ bib | .pdf ] |
[14] | P. Irofti and B. Dumitrescu, “Overcomplete Dictionary Learning with Jacobi Atom Updates,” in 39th International Conference on Telecommunications and Signal Processing, 2016, pp. 421--424. [ bib | .pdf ] |
[15] | P. Irofti and B. Dumitrescu, “Regularized Algorithms for Dictionary Learning,” in 2016 International Conference on Communications (COMM), 2016, pp. 439--442. [ bib | .pdf ] |
[16] | P. Irofti, “Sparse Denoising with Learned Composite Structured Dictionaries,” in 19th International Conference on System Theory, Control and Computing, 2015, pp. 331--336. [ bib | .pdf ] |
[17] | P. Irofti and B. Dumitrescu, “Cosparse Dictionary Learning for the Orthogonal Case,” in 19th International Conference on System Theory, Control and Computing, 2015, pp. 343--347. [ bib | .pdf ] |
[18] | P. Irofti and B. Dumitrescu, “Overcomplete Dictionary Design: the Impact of the Sparse Representation Algorithm,” in The 20th International Conference on Control Systems and Computer Science, 2015, pp. 901--908. [ bib | .pdf ] |
[19] | P. Irofti and B. Dumitrescu, “GPU Parallel Implementation of the Approximate K-SVD Algorithm Using OpenCL,” in 22nd European Signal Processing Conference, 2014, pp. 271--275. [ bib | .pdf ] |
Workshops and conferences with extended abstract:
[1] | P. Irofti and A. Băltoiu, “Unsupervised Dictionary Learning for Anomaly Detection,” in International Traveling Workshop on Interactions Between Sparse Models and Technology, 2020, pp. 1--3. [ bib | arXiv ] |
[2] | P. Irofti and F. Stoican, “Dictionary Learning Applications in Control Theory,” in 1st Conference on Recent Advances in Artificial Intelligence, 2017, pp. 1--2. [ bib | .pdf ] |
[3] | P. Irofti, “The Effect of Atom Replacement Strategies on Dictionary Learning,” in International Traveling Workshop on Interactions Between Sparse Models and Technology, 2016, pp. 10--11. [ bib | .pdf ] |
[4] | P. Irofti, “Making OpenBSD Useful on the Octeon Network Gear,” in European BSD Conference, 2014. [ bib | .pdf ] |
[5] | P. Irofti, “Dynamic Analysis: Knowing When to Stop,” in Challenges in Analysing Executables: Scalability, Self-Modifying Code and Synergy (Dagstuhl Seminar 14241), 2014. [ bib | .pdf ] |
[6] | P. Irofti, “Porting OpenBSD on the MIPS64-based Octeon Platforms,” in BSDCan Conference, 2014. [ bib | .pdf ] |
[7] | P. Irofti, “Emulator Design, Traps and Pitfalls,” in Analysis of Executables: Benefits and Challenges (Dagstuhl Seminar 12051), 2012. [ bib | .pdf ] |
[8] | P. Irofti, “OpenBSD's New Suspend and Resume Framework,” in European BSD Conference, 2011. [ bib | .pdf ] |
[9] | P. Irofti, “On the Linux Compatibility Layer in OpenBSD 5.0,” in Slackathon Conference, 2011. [ bib | .pdf ] |
Preprints:
[1] | P. Irofti, A.I. Hîji, A. Pătrașcu, and N. Cleju, “Fusing Dictionary Learning and Support Vector Machines for Unsupervised Anomaly Detection,” pp. 1--35, 2024. [ bib | arXiv ] |
[2] | A. Pătrașcu, C. Rusu, and P. Irofti, “Learning Explicitly Conditioned Sparsifying Transforms,” pp. 1--18, 2024. [ bib | arXiv ] |
[3] | A. Pătrașcu and P. Irofti, “Computational complexity of Inexact Proximal Point Algorithm for Convex Optimization under Holderian Growth,” pp. 1--42, 2021. [ bib | arXiv ] |
Theses:
[1] | P. Irofti, Parallel Dictionary Learning Algorithms for Sparse Representations, Ph.D. thesis, University Politehnica of Bucharest, 2015. [ bib | .pdf ] |
[2] | P. Irofti, “Image Processing with Wavelet Transformations,” 2008, Bachelor of Science Thesis in Romanian. [ bib | .pdf ] |