UE Advanced algorithms for machine learning and data mining

Diplômes intégrant cet élément pédagogique :

Descriptif

This lecture introduces some of the most common advanced algorithms of machine learning and data mining, from the computation of similarities in high-dimensional spaces to backpropagation in recurrent neural networks and soft, kernel and spectral clustering. The models behind these algorithms will be detailed as well as their main applications pertaining to text, graph and sequence mining. We will more particularly study the following topics:

  1. Computing similarities with high-dimensional, sparse representations
  2. Random walks and the PageRank algorithm
  3. Ranking problems and learning to rank
  4. Introduction to recurrent neural networks and natural language processing
  5. Probabilistic estimation: an introduction to MCMC and Gibbs sampling
  6. Batch and online partitioning/hierarchical clustering
  7. Soft and fuzzy clustering
  8. Constrained clustering
  9. Kernel and Spectral clustering
  10. Sparse coding and dictionary learning for clustering

Pré-requis

Fundamentals of probability/statistics, linear algebra and computer science (data structures and algorithms)

Bibliographie

C.D. Manning, P. Raghavan and H. Schütze. Introduction to Information Retrieval. Cambridge University Press, USA, 2008.

A. DasGupta. Probability for Statistics and Machine Learning. Springer, 2011.

I. Goodfellow, Y. Bengio, A. Courville. Deep Learning. MIT Press, 2016.

C.M. Bishop. Pattern Recognition and Machine Learning. Springer Verlag, 2006.

Informations complémentaires

Méthode d'enseignement : En présence
Lieu(x) : Grenoble - Domaine universitaire
Langue(s) : Anglais