Ahmed El Alaoui

Learn More
One approach to improving the running time of kernel-based machine learning methods is to build a small sketch of the input and use it in lieu of the full kernel matrix in the machine learning task of interest. Here, we describe a version of this approach that comes with running time guarantees as well as improved guarantees on its statistical performance.(More)
One approach to improving the running time of kernel-based methods is to build a small sketch of the kernel matrix and use it in lieu of the full matrix in the machine learning task of interest. Here, we describe a version of this approach that comes with running time guarantees as well as improved guarantees on its statistical performance. By extending the(More)
Given a weighted graph with N vertices, consider a real-valued regression problem in a semisupervised setting, where one observes n labeled vertices, and the task is to label the remaining ones. We present a theoretical study of `p-based Laplacian regularization under a d-dimensional geometric random graph model. We provide a variational characterization of(More)
We propose to measure the Beam Spin Asymmetry (BSA) in Deeply Virtual Compton Scattering (DVCS) off 4He. The measurements will use a 6 GeV polarized electron beam, a 4He pressurized gas target, the CLAS and the BoNuS RTPC detectors. The major goal of this proposal is to perform the first fully quantitative investigation of the DVCS reaction off a nuclear(More)
We present new algorithms for computing the log-determinant of symmetric, diagonally dominant matrices. Existing algorithms run with cubic complexity with respect to the size of the matrix in the worst case. Our algorithm computes an approximation of the log-determinant in time near-linear with respect to the number of non-zero entries and with high(More)
Consider a population consisting of n individuals, each of whom has one of d types (e.g. their blood type, in which case d = 4). We are allowed to query this database by specifying a subset of the population, and in response we observe a noiseless histogram (a d-dimensional vector of counts) of types of the pooled individuals. This measurement model arises(More)
We consider the problem of decoding a discrete signal of categorical variables from the observation of several histograms of pooled subsets of it. We present an Approximate Message Passing (AMP) algorithm for recovering the signal in the random dense setting where each observed histogram involves a random subset of size proportional to n of entries. We(More)
Measuring Deeply Virtual Compton Scattering on a neutron target is one of the necessary steps to complete our understanding of the structure of the nucleon in terms of Generalized Parton Distributions (GPDs). DVCS on a neutron target allows to operate a flavor decomposition of the GPDs and plays a complementary role to DVCS on a transversely polarized(More)
High-statistics measurements of differential cross sections and recoil polarizations for the reaction γp → K + 0 have been obtained using the CLAS detector at Jefferson Lab. We cover center-of-mass energies (√ s) from 1.69 to 2.84 GeV, with an extensive coverage in the K + production angle. Independent measurements were made using the K + pπ − (γ) and K +(More)
A. Kimb,a,∗, H. Avakian c, V. Burkert c, K. Joo a, W. Kimb, K.P. Adhikari e,d, Z. Akbar f, S. Anefalos Pereira g, R.A. Badui h, M. Battaglieri i, V. Batourine c, I. Bedlinskiy j, A.S. Biselli k, S. Boiarinov c, P. Bosted l,c, W.J. Briscoem, W.K. Brooks n, S. Bültmann e, T. Cao o, D.S. Carman c, A. Celentano i, S. Chandavar p, G. Charles q, T. Chetry p, L.(More)