Moez DRAIEF

Senior Lecturer - Associate Professor

 

Short Bio


I graduated from the Ecole Polytechnique (Paris) in 2000. I then completed a DEA in Probability Theory at the University Paris VI. I undertook a PhD at the LIAFA (Theoretical Computer Science Group), University Paris VII. 


From October 2004 to January 2007, I was a Marie Curie research fellow at the Statistical Laboratory and a lecturer in Part III (Certificate of Advanced Study in Mathematics), Cambridge University.


Since 2007, I am a Faculty member in the EEE department, Imperial College London.


I spent the first half of 2011 as a visiting researcher to Microsoft Research Cambridge visiting the Networks, Economics and Algorithms group.


In Fall 2014, I will be teaching, with Wei Dai, a course Topics in Large Dimensional Data Processing.


In 2014, I co-founded, with. A. Mijatovic, the Imperial Probability Centre.


I am currently on sabbatical at Huawei Technologies, Mathematical and Algorithmic Sciences Lab, based in Paris, where I lead the Computational Learning, Algorithms and Networks Group.


Research Interests


My main research interests are the analysis of distributed algorithms, and network economics and dynamics as well as machine learning in the context of SDN and 5G wireless systems.

It involves the mathematical analysis of models of large networks that arise in online and communications systems  using tools from probability theory, statistical learning and algorithm design.


Here are a few papers highlighting my recent interests.


Online and Golbal Network Optimzation:  Towards the enxt generation of routing platforms

J. Leguay, M. Draief, S. Chouvardas, L. Maggi, S. Paris, G. Paschos and M. Qi

Proc. IEEE NOMS 2016.


Admission Control with Online Algorithms in SDN

J. Leguay, S. Chouvardas, L. Maggi, S. Paris and M. Draief

Proc. IEEE NOMS 2016.


A diffusion Kernel LMA algorithm for non-linear adaptive networks

S. Chouvardas and M. Draief

Proc. IEEE ICASSP 2016.


A reduced path loss reconstruction method using matrix completion and adaptive samplings

S. Chouvardas, S. Valentin, M. Draief and M. Leconte

Proc. IEEE ICASSP 2016.


Placing dynamic content in caches with small populations

M. Leconte, G. Paschos, S. Chouvardas, M. Draief et al

Proc. IEEE INFOCOM 2016.


Global Majority Consensus by Local Majority Polling

  1. M.Abdullah and M. Draief

Discrete Applied Mathematics, 2015


Viral Processes by Random Walks on Random Regular Graphs

M. Abdullah, C. Cooper and M. Draief

Annals of Applied Probability, 2015


On the Power of Adversarial Infection in Networks

M. Brautbar, M. Draief and S. Khanna

Internet Mathematics, 2015.


Large-Scale Bandit Problems and KWIK Learning

  1. K.Amin, J. Abernethy M. Draief and M. Kearns

Journal of Machine Learning Research, Proc. ICML 2013.


Flooding in Weighted Sparse Random Graphs

  1. H.Amini, M. Draief and M. Lelarge

SIAM Journal on Discrete Mathematics, 2013.


Convergence Speed of Binary Interval Consensus

M. Draief and M. Vojnovic

SIAM Journal Control and Optimization, 2012.


See below for my other contributions.


Book “Epidemics and Rumours in Complex Networks”, Cambridge University Press.


With L. Massoulié, I co-authored a book “Epidemics and Rumours in Complex Networks”.




Here is the table of content with pdf files of some chapters.


0- Introduction

1- Galton-Watson branching processes

2- Reed-Frost epidemics and Erdos-Renyi graphs

3- Connectivity and Poisson approximation

4- Diameter of Erdos-Renyi graphs

5- From microscopic to macroscopic dynamics

6- The small-world phenomenon

7- Power laws via preferential attachment

8- Epidemics on general graphs

9- Viral Marketing and optimised epidemics



Selected Papers



Optimal dynamic immunisation under a controlled heterogeneous node-based SIRD model

L-X Yang, M. Draief and M. Yang

PHYSICA A, to appear.


Probabilictis Abstract Argumentation : an investigation with Boltzmann machines

  1. R.Riveret, J. Pitt, M. Draief and D. Kornikov.

Proc. AAMAS 2015.

Argumentation and Computation Journal 2015.


Impact of the network topology on the viral prevalence:  a node based-approach

L-X Yang, M. Draief and M. Yang

PLoS One, 2015.


Speeding up Cover Time of Sparse Graphs using Local Knowledge

  1. M.Abdullah, C. Cooper and M. Draief

Proc. IWOCA 2015.


Opinion Formation Models on a Gradient

  1. M.Gastner, N. Markou, G. Pruessner and M. Draief

PLoS One, 2014.


New Models for Competitive Contagion

M. Draief, H. Heidari and M. Kearns

Proceedings of AAAI 2014.


Global Majority Consensus by Local Majority Polling

  1. M.Abdullah and M. Draief

Proc. NetGCoop 2014.

Discrete Applied Mathematics, 2015


Viral Processes by Random Walks on Random Regular Graphs

M. Abdullah, C. Cooper and M. Draief

Proceedings of RANDOM 2011.

Annals of Applied Probability, 2015


On the Power of Adversarial Infection in Networks

M. Brautbar, M. Draief and S. Khanna

Proc. Algorithms and Models for the Web Graph (WAW’13).

Internet Mathematics, 2015.


Large-Scale Bandit Problems and KWIK Learning

  1. K.Amin, J. Abernethy M. Draief and M. Kearns

Journal of Machine Learning Research, Proc. ICML 2013.


Structural Analysis of Viral Spreading Processes using Egonets

  1. V.Preciado, M. Draief and A. Jadbabaei


Distributed Multivalued Consensus

A. Babaee and M. Draief

Computer Journal, 2014.


Contagion and Observability in Security Domains

  1. Y.Bachrach, M. Draief and Sanjeev Goyal

Proc. Allerton Communication Control and Computing (Allerton'13).


Flooding in Weighted Sparse Random Graphs

  1. H.Amini, M. Draief and M. Lelarge

Proc. SIAM Analytic Algorithms and Combinatorics (ANALCO'11).

SIAM Journal on Discrete Mathematics, 2013.


Convergence Speed of Binary Interval Consensus

M. Draief and M. Vojnovic

Proceedings of Infocom, 2010

SIAM Journal Control and Optimization, 2012.


UKAIRO: Internet-Scale Bandwidth Detouring

  1. T.Haddow, SW. Ho, C. Lumezanu, M. Draief and P. Pietzuch


On the Feasibility of Bandwidth Detouring

T. Haddow, S.W. Ho, J. Ledlie, C. Lumezanu, M. Draief, and P. Pietzuch,

Proceedings of Passive and Active Measurement Confrence (PAM) 2011.


Interleaving Multi-Agent Systems and Social Networks for Organized Adaptation

J. Pitt, D. Ramirez-Cano, M. Draief, and A. Artikis

Proc. SNAMAS 2009.

Computational and Mathematical Organization Theory, 2011.


Bargaining Dynamics in Exchange Networks

  1. M.Draief and M. Vojnovic

Proc. Allerton Communication Control and Computing (Allerton'10).


Percolation and competing innovations

C. Roca, M. Draief and D. Helbing

in Understanding Complex Systems: Social Self-Organization

Ed. D. Helbing, Ch. 9. “Coordination and Competitive Innovation Spreading in Social Networks”

p.169-184, 2012.


A random walk model for infection on graphs

M. Draief and A. Ganesh

Proc. Valuetools 2009.

Discrete Event Dynamic Systems, 2010.


Marketing in a random network

H. Amini, M. Draief, and M. Lelarge  

Proc. Network Control and Optimization, 2009.


Exponential random graphs as models of overlay networks

M. Draief, A. Ganesh, and L. Massoulié

Journal of Applied Probability, 2009.


Deconstructing Internet Paths-  An Approach for AS-Level Detour Route Discovery

  1. S.W. Ho, T. Haddow, J. Ledlie, M. Draief, and P. Pietzuch

Proceedings of the 8th International Workshop on Peer-to-Peer Systems (IPTPS), 2009.


Thresholds for virus spread on networks

  1. M.Draief, A. Ganesh, and L. Massoulié

Proc. Valuetools 2006.

Annals of Applied Probability, 2008.


The single server queue and the storage model- Large deviations and fixed points

M. Draief

International Journal of Stochastic Analysis, 2006.


Efficient routeing in Poisson small-world networks

M. Draief and A. Ganesh

Journal of Applied Probability, 2006.


Queues, stores, and tableaux 
M. Draief, J. Mairesse, and N. O'Connell

Journal of Applied Probability, 2005.


Services within a busy period of an M/M/1 queue and Dyck paths

M. Draief and J. Mairesse

Queueing Systems, 2005.