Paul Irofti


About me: 
 Resume (RO)
 Experience
 Research
 Education
 Papers
 Notes
 Hobbies
 LinkedIn

Services: 
 Training
 Development
 System Administration
 Security Audits

Contact: 
 [E-mail address]

Current Position

Department of Automatic Control and Computers
University Politehnica of Bucharest

Postdoc involved in two research projects. I am also teaching "Numerical Methods", "Parallel and distributed processing of data and knowledge", and "Modelling and Simulation".

Sparse representations in signal processing
IDEI 2011 project, January 2014 - Present

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 - Present

Investigating sensor placement using sparse representations and dictionary learning with applications in water distribution networks.

Publications

Journal:

[1] B. Dumitrescu and P. Irofti, “Regularized K-SVD,” IEEE Signal Processing Letters, vol. 24, no. 3, pp. 309--313, March 2017. [ bib | DOI | .pdf ]
[2] 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 ]
[3] 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] 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 ]
[2] 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 ]
[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 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 ]
[5] P. Irofti and B. Dumitrescu, “Regularized Algorithms for Dictionary Learning,” in 2016 International Conference on Communications (COMM), 2016, pp. 439--442. [ bib | .pdf ]
[6] 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 ]
[7] 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 ]
[8] 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 ]
[9] 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 ]

Under Review:

[1] B. Dumitrescu and P. Irofti, Algorithms and Applications for Dictionary Learning, Springer, 2017. [ bib ]

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 ]