Algorithms and their decision making

Increasingly, algorithms are replacing human beings in decision-making or the performance of tasks. They take into account a number of parameters, but the question of the transparency of the algorithms arises: can they explain their decision?

Gradually, we let the algorithms take over from our decisions. They are used for example by banks to obtain a loan, or to replace a judge. The banker can have a grievance against the loan applicant, or use a software biased. The judge, on the other hand, does not always detail the reasons that guided his decision. A algorithm is above all a method to find a solution to a problem, it can be fairer, have fewer prejudices unlike humans, it has no preconceptions and at least does not change mood. Human justice has faults, humans can be biased. Ethnicity, media context … so many circumstances that can influence judges. A new concept is emerging: that of the quality of data, and know how to define what is good data.

Algorithms make decisions, but the motivation behind them cannot be theirs alone. For their decisions to be fully accepted, learning algorithms who could explain their result.