31 reputation
6
bio website vision.ai.uiuc.edu/~bghanem2
location Singapore
age
visits member for 4 years, 2 months
seen Apr 6 '12 at 10:40
Interested in research topics related to computer vision and machine learning, especially large scale optimization and randomization techniques for classification.

Sep
24
awarded  Autobiographer
Jul
2
awarded  Curious
Feb
2
awarded  Popular Question
Mar
6
awarded  Popular Question
Apr
6
revised conjugate function for matrix mixed norm
added 1 characters in body; edited body
Apr
3
asked conjugate function for matrix mixed norm
Mar
29
comment Projected gradient descent for non-convex optimization problems
Thanks Brian for the comments!
Mar
27
asked Projected gradient descent for non-convex optimization problems
Jan
7
comment smooth approximation of the hinge loss function
Sorry about the link. The paper can be viewed here: cs.uiuc.edu/homes/afarhad2/index_files/scene_discovery.pdf. Actually, I found the proper coefficients for the order 4 polynomial. The paper has a typo on page 5. Instead of $\frac{3}{2}x$ it should be $\frac{1}{2}x$. The link for the smooth quadratic hinge function was a great help. Thanks Suvrit.
Jan
7
asked smooth approximation of the hinge loss function
Dec
27
comment Statistical estimation of singular values and vectors
@ J.M.: The matrix is an error matrix that varies in a complex way. There is dependence on other variables, so I can't model how the variation goes. I can say that the change is small, so can you refer me to the SVD update algorithms you had in mind? Thanks
Dec
27
comment Statistical estimation of singular values and vectors
@ J.M.: The randomization techniques I referred to show that they are more efficient than the SVD power algorithm mentioned in the papers you suggested.
Dec
27
comment Statistical estimation of singular values and vectors
Let me rephrase the question. Given an mxn matrix A, I need to compute the largest k singular values and vectors (U, $\Sigma$ and V). Are there any available randomized techniques that can estimate these values and vectors, whose computational complexity is better than the deterministic technique of SVD? I found this paper: cims.nyu.edu/~tygert/randsurvey.pdf. It shows 2 algorithms that can estimate the SVD but indirectly. It does so by first performing a randomized interpolative decomposition (ID). Does anyone know of a randomized technique that estimates SVD directly?
Dec
27
asked Statistical estimation of singular values and vectors
Oct
15
awarded  Editor
Oct
15
comment Maximum average value within a rectangular bounding box
i updated the question, so any extra comments would be appreciated. Thanks!
Oct
15
comment Maximum average value within a rectangular bounding box
i updated the question, so any extra comments would be appreciated. Thanks!
Oct
15
revised Maximum average value within a rectangular bounding box
added 495 characters in body
Oct
14
awarded  Student
Oct
14
asked Maximum average value within a rectangular bounding box