Research interests: High-dimensional phenomena in statistics and probability theory. Related algorithmic questions. Statistical physics.
The questions that drive my research are the fundamental limits of extracting information from noisy data, and the algorithmic feasibility of this task. I like to think about large random structures such as matrices, graphs and tensors, and understand how sudden changes in their structural properties have statistical and algorithmic consequences.
Currently teaching (spring 2021) : STSCI6940 Topics in high-dimensional inference.
Article about me by Chris Dawson for Cornell Engineering.
Email: elalaoui [at] cornell [dot] edu
- Algorithmic pure states for the negative spherical perceptron. A. El Alaoui, M. Sellke. Preprint 2020. [arxiv].
- Algorithmic thresholds in mean-field spin glasses. A. El Alaoui, A. Montanari. Preprint 2020. [arxiv].
- Imputation for high-dimensional linear regression. K. Chandrasekher, A. El Alaoui, A. Montanari. Preprint 2020. [arxiv].
- On the computational tractability of statistical estimation on amenable graphs. A. El Alaoui, A. Montanari. Submitted 2019. [arxiv].
- Finite-size corrections and likelihood ratio fluctuations in the spiked Wigner model. A. El Alaoui, F. Krzakala, M. I. Jordan. Preprint 2017. [arxiv].
(Working paper; will be substantially revised.)
- Optimization of mean-field spin glasses. A. El Alaoui, A. Montanari, M. Sellke. Annals of Probability (to appear). [arxiv].
- Fundamental limits of detection in the spiked Wigner model. A. El Alaoui, F. Krzakala, M. I. Jordan. Annals of Statistics, Vol 48, No. 2, 863-885, 2020. [journal, arxiv].
- Decoding from pooled data: Sharp information-theoretic bounds. A. El Alaoui, A. Ramdas, F. Krzakala, L. Zdeborová, M. I. Jordan. SIAM Journal on Mathematics of Data Science 1-1 (2019), pp. 161-188. [journal, arxiv].
- Decoding from pooled data: Phase transitions of message passing. A. El Alaoui, A. Ramdas, F. Krzakala, L. Zdeborová, M. I. Jordan. IEEE Transactions on Information Theory, 65, 572-585, 2019. [journal, arxiv].
Presented at IEEE International Symposium on Information Theory (ISIT) 2017.
- The Kikuchi hierarchy and tensor PCA. A. Wein, A. El Alaoui, C. Moore. 60th Annual Conference on Foundations of Computer Science (FOCS) 2019. [arxiv].
- Detection limits in the high-dimensional spiked rectangular model. A. El Alaoui, M. I. Jordan. 31th Annual Conference on Learning Theory (COLT), PMLR 75:410-438, 2018. [proc., arxiv].
- Tight query complexity lower bounds for PCA via finite sample deformed Wigner law. M. Simchowitz, A. El Alaoui, B. Recht. 50th Annual Symposium on the Theory of Computing (STOC) 2018. [proc., arxiv].
Here’s an earlier version (not intended for publication) with slightly suboptimal results.
- Estimation in the spiked Wigner model: A short proof of the replica formula. A. El Alaoui, F. Krzakala. IEEE International Symposium on Information Theory (ISIT) 2018. [proc., arxiv].
- Asymptotic behavior of Lp-based Laplacian regularization in semi-supervised learning. A. El Alaoui, X. Cheng, A. Ramdas, M. J. Wainwright, M. I. Jordan. 29th Annual Conference on Learning Theory (COLT), PMLR 49:879-906, 2016. [proc., arxiv].
- Fast randomized kernel ridge regression with statistical guarantees. A. El Alaoui, M. W. Mahoney. Advances in Neural Information Processing Systems (NIPS) 28, 2015. [proc., arxiv].
- CS174 Combinatorics and Discrete Probability (UC Berkeley, spring 2015).
- STAT210B Theoretical Statistics, part B (UC Berkeley, spring 2017).
- (2021-) Assistant professor, Department of Statistics and Data Science, Cornell University.
- (Fall 2020) Richard M. Karp Research Fellow, Simons Institute for the Theory of Computing, UC Berkeley.
- (2018-2020) Postdoctoral researcher, Stanford University. Hosted by Andrea Montanari.
- (2013-2018) Ph.D. Electrical Engineering and Computer Sciences, UC Berkeley. Advised by Michael I. Jordan.
- (2012-2013) M.Sc. Mathématiques, Vision et Apprentissage, Ecole Normale Supérieure/Ecole des Ponts Paristech.
- (2009-2012) Eng.Deg. Applied math, Ecole Polytechnique.
- Detection limits and fluctuation results in some spiked random matrix models and pooling of discrete data. [link].
Ahmed El Alaoui.
Ph.D. Thesis, Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, 2018.