Mouhamadou Moustapha Cisse
Universit Pierre et Marie Curie (UPMC)
e
Laboratoire Informatique Paris 6 (LIP6)
Citizenship : Senegalese
Phone number (mobile) : +33 6 24 67 26 94
E-mail : *****@******.****.**
Education
October 2010 - Present PhD candidate in machine learning advised by Professor Pa-
trick Gallinari and Professor Thierry Artieres Expected gra-
duation on spring 2014 Machine learning department of LIP6
2008–2010 Master in computer science Artificial intelligence and decision theory
specialization (with honours) University of Pierre et Marie Curie
(Paris)
2007–2008 Master (Maitrise) in software engineering (with honours) Uni-
versity Gaston Berger (Senegal)
2004–2007 Bachelor in Mathematics and computer science (with honours)
University Gaston Berger (Senegal)
Research experience/Projects
Since October 2010 PhD candidate in machine learning advised by Professor Pa-
trick Gallinari and Professor Thierry Arti`res We work on
e
reducing the inference complexity of extreme classification. We have
proposed solutions for both the single and the Multilabel cases. Ma-
chine learning department LIP6-UPMC
Summer 2010 Research internship supervised by Professor Thierry Artieres
We worked on Training time reduction of deep neural networks by
using random projections. Machine learning department UPMC-
LIP6
Winter 2010 Research project in machine learning supervised by Professor
Yoshua Bengio (exchange semester) We improved Handwrit-
ten character recognition by using out of sample distribution data
in unsupervised training of deep neural networks (work published in
JMLR). Machine learning department (LISA, University of Mon-
treal)
Summer 2009 Research internship supervised by Professor Ryutaro Ichise Ap-
plied machine learning methods for semantic integration of structu-
red documents using linked data information. National Institute of
informatics (Tokyo-Japan)
Publications
2013 Robust Bloom Filters for Large Multilabel Classification Tasks
Moustapha Ciss , Nicolas Usunier, Thierry Arti`res, Patrick Gallinari
e e
Neural Information Processing Systems (NIPS), 2013 Lake Tahoe,
USA
2012 Learning Compact Class Codes for Fast Inference in Large
Multi Class Classification
Moustapha Ciss , Thierry Arti`res, Patrick Gallinari
e e European
Conference on Machine Learning (ECML/PKDD), 2012 Bristol, UK
2011 Deep Learners Benefit More from Out-of-Distribution
Examples
Yoshua Bengio, et al. AISTATS 2011 - Journal of machine learning
research C&WP Florida, USA
2010 Deep Self Taught Learning for Handwritten Digit Recognition
Yoshua Bengio, et al. NIPS 2010 Deep Learning Workshop Van-
couver, Canada
Research interests
Machine Learning, Extreme Classification, Algorithms, Deep Neural Networks, Embedding, Information
Retrieval.
Teaching(UPMC)
Bachelor level : Li215 Advanced programming in C
Li214 Algorithms and data structures
Li345 Database management systems and XML
Programming Skills
Programming languages : Python, C/C++
Modelling : UML.
Databases and knowledge representation : SQL3, XML
Scientific Tools : Numpy/Scipy, Matlab
Fluency in English, French and Wolof (Senegalese), Spanish (functional)
Others
2012 IPAM Deep Learning Graduate Summer School
UCLA, California-USA
2011 Machine Learning Summer School
Bordeaux, France
2010-2013 Merit Based French Ministry of Research Scolarship For PhD
studies
2009-2010 Merit Based Senegalese Government Scolarship for Graduate
Studies