Join the Mooc "From probabilities to Bayesian estimation"

This 7-week course (in French) is mainly designed for master's students in engineering schools.

Everyday life is characterised by happenstance:
 

• we do not always take the same amount of time between home and work;
• a heavy smoker may or may not suffer from cancer;
• the fish don't always bite.

Such phenomena are known as random or stochastic. Quantifying them naturally leads to the use of probability theory.

In the smoking example, let's imagine that the doctor does not trust his patient's statements about his cigarette consumption. He decides to have his blood nicotine level measured by a medical laboratory. Probability theory offers us tools to quantify the stochastic link between the number of cigarettes per day and the nicotine level.

From this nicotine level, we will be able to estimate the number of daily cigarettes. Estimation theory offers several solutions:
 

• the most likely;
• the most probable;
• the average value.

These seemingly similar concepts have very different meanings in estimation theory. We will distinguish between classical estimation and Bayesian estimation.

At the end of this course:
 

• you will have understood that there is no magic algorithm that would allow problems such as those mentioned above to be solved;
• you will be able to ask the specialist in the field to develop a model linking the quantities to be estimated to the observed quantities;
• you will be able to develop an estimation algorithm to reconstruct the quantities to be estimated from the observed quantities.

► Register on Fun Mooc from 28 February

MOOC team

• Éric Le Carpentier, lecturer at Centrale Nantes
• Julie Grosclaude, educational engineer at Atlanstic 2020
• Vanessa Le Garrec, educational engineer at Centrale Nantes
• Dominika Jankosikova, educational engineer at Centrale Nantes
• Patrick Roustang, videographer at Centrale Nantes
• Romain Plourde, videographer at Centrale Nantes