3
Machine Intelligence – Evaluation Functions and Numerical Convergence
Two issues appear when a reward matrix (R)-driven MDP produces results. These issues can be summed up in two principles.
Principle 1: AI algorithms often surpass humans in classification, prediction, and decision-making areas.
The key executive function of human intelligence, decision-making, relies on the ability to evaluate a situation. No decision can be made without measuring the pros and cons and factoring the parameters.
Humanity takes great pride in its ability to evaluate. However, in many cases, a machine can do better. Chess represents our pride in our thinking ability. A chessboard is often present in movies to symbolize human intelligence.
Today, not a single ...
Get Artificial Intelligence By Example - Second Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
