Probability as degree of confidence

When developers hear the word "probability", they often imagine dice, coin flips, and the school formula "favorable outcomes divided by all possible outcomes". This is useful, but a very narrow picture. In machine learning and applied analytics, probability almost always means something else – the degree of our confidence in a statement given the available data.

• Probability example with softmax
• Case 1. Spam filter: probability ≠ decision
• Case 2. Medical test: updating confidence
• Case 3. Multiclass classification and softmax
• Case 4. Overconfident model as a problem signal
• Case 5. Updating confidence with new data