Advanced Deep Learning with Python

Conditional probability and the Bayes rule

If the occurrence of event A changes the probability of the occurrence of event B, where A occurs before B, then the two are dependent. To illustrate this concept, let's imagine that we draw multiple cards sequentially from the deck. When the deck is full, the probability to draw hearts is P(hearts) = 13/52 = 0.25. But once we've drawn the first card, the probability to pick hearts on the second turn changes. Now, we only have 51 cards and one less heart. We'll call the probability of the second draw conditional probability and we'll denote it with P(B|A). This is the probability of event B (second draw), given that event A has occurred (first draw). To continue with our example, the probability ...