The most understated truth in Data Science. Indeed designing ones own algorithm based on the raw building blocks of uncertain reasoning, that is probability theory, logic and information theory, will not only give one a better understanding of the domain but in general results in more accurate models.

]]>