Teaching Activities at IMT Lille Douai

Please find below a description of the lectures in which I am mostly involved.


DATA Science (M2) [Coordinator]

I have proposed and created in 2016 this course (120h approx.) - The topics covered here include:
  • Statistical Methods for Machine Learning:
    • Regression Problems
    • Bayesian Methods
    • Clustering and Mixture Models
    • Neural Networks
    • Classification
    • Large Margin Classifiers Decision Tree
    • Ensemble Methods
  • Monte Carlo Methods
  • Convex Optimization
  • Times Series: hidden Markov model, Kalman filter and Particle Filtering
  • Seminars from professional: Dataiku, Vekia, etc.
  • Data Science Project (Challenge with real dataset on the “Kaggle inclass” platform)

Statistical Inference (L3) [Coordinator]

The objective is to introduce the following concepts:
  • Hypothesis testing: definition, exhaustivity, Likelihood ratio, Neyman-Pearson, Bayes criterion,
  • Parametric estimation: estimator, variance / bias, Fisher information, cramer-Rao bound, etc.

Signal Processing (L3) [Coordinator]

The objective is to introduce the following concepts:
  • Deterministic signals:
    • Classification of signals: finite energy, finite power, periodic
    • Convolution and its properties; Delta functions and their properties.
    • Fourier Analysis: Fourier Series and Transform
    • Linear filtering: impulse response, stability, ideal filters (lowpass, highpass and band-pass)
  • Digital signals: Filtering (structure, stability, causality), Z-transform and its inverse, discrete Fourier transform
  • Random signals:
    • Brief course reminders on Probability: random variables, distributions, common distributions.
    • Ergodicity and stationarity
    • Spectral analysis
    • Filtering of stochastic process.

Probability (L3)

The objective of this course is to provide the basic principle for the analysis of random variables:
  • Probability space
  • Discrete and continuous random variables: cumulative distribution function, probability density function, expectation, change of variables
  • Common probability density functions: uniform, Bernoulli, binomial, Poisson, normal, exponential,...
  • Joint distributions, marginal and conditional distributions, covariance, correlation, independence