I am currecntly developing algorithms for solving various problems in machine learning, machine vision, computational neuroscience and speech processing. Developed algorithms are based on advanced procedures in optimisation (like manifold optimisation), signal processing (like wavelet) and statistics (like information criteria).
- Ph.D., Computer Science (Machine Learning), Technical University of Berlin, 2007-2012.
- M.Sc., Biomedical Engineering, Amirkabir University, 2004-2006
- B.Sc., Electrical Enginnering , Tehran University, 2000-2004
- Recent Publication:
1- Books/Book Chapters:
- Suvrit Sra, Reshad Hosseini. Geometric Optimization in Machine Learning. In Algorithmic Advances in Riemannian Geometry and Applications, Springer, 2016. (more...)
- Pourya H. Zadeh, Reshad Hosseini. Expected Logarithm of Central Quadratic Form and Its Use in KL-Divergence of Some Distributions. Entropy, Aug 2016. (more...)
- Reshad Hosseini, Suvrit Sra, Lucas Theis, Matthias Bethge. Inference and mixture modelling with the Elliptical Gamma Distribution. Computational Statistics and Data Analysis (CSDA), Feb 2016. (more...)
- Arash Mehrjou, Reshad Hosseini, Babak Nadjar Araabi, Improved Bayesian information criterion for mixture model selection. Pattern Recognition Letters, 2016. (more...)
- Pourya H. Zadeh, Reshad Hosseini, Suvrit Sra. Geometric Mean Metric Learning. International Conference on Machine Learning (ICML 2016) (more...)
- Reshad Hosseini, Suvrit Sra. Matrix Manifold Optimization for Gaussian Mixtures. Advances in Neural Information Processing Systems (NIPS 2015) (more...)
Suvrit Sra, Reshad Hosseini, Lucas Theis, Matthias Bethge. Data Modeling with the Elliptical Gamma Distribution. Artificial Intelligence and Statistics (AISTATS 2015) (more...
- More Publication:
- Principal Investigator, Cheque Authentication, Informatic Service Co., 2016.
- Reseacher, Image Compression, White Matter Technology Co., 2010.
- Professinal Activities/Experinces:
- Granted patent titled " Method and Device for Image Compression", US Patent Office, 2014.
- Granted patent titled "Method and Device for Image Compression", European Patent Office, 2013.
- Granted patent titled " Image Compression Process through Transforming an Image Block based on Specific Transfer Assigned to that Class", Iraninan Patent Office, 2013.
- “New estimate for the redundancy of natural images”, Bernstein Conference of Computational Neuroscience, 2010.
- “Spectral Stacking: Unbiased shear estimation for weak gravitational lensing”, COSMOSTAT09, 2009.
- Honors and Awards:
- Membership in National Elite Foundation, 2015.
- GREAT08 Challenge Award for Presentation of the Best Method in GREAT08 PASCAL Contest, Gravitational Lensing and Accuracy Testing, 2008