Probalistic Reasoning

Thoughts on probabilistic reasoning

In the situation of deduction, the system is able to draw conclusions from general rules or premises. This is a deterministic process, meaning that the system will always reach the same conclusion given the same input. Also, we consider that each of the steps is 100% correct, so the chaining of the steps will also be 100% correct.

However, in the real world, we have to deal with uncertainty, and the system has to be able to handle it. This is where probabilistic reasoning comes into play.

Probabilistic reasoning allows to assign probabilities to the different steps of the reasoning process. This can be useful in several situations:

• When the system is not sure about the correctness of a step, for instance because it was guessed from unstructured data.
• When the system has to deal with incomplete information, and has to make assumptions.
• When the system has to deal with conflicting information.

In those situations, the system can assign probabilities to the different steps of the reasoning process, and then use those probabilities to compute an aggregate probability of the final conclusion.