Getting Started with Next-Generation Artificial Intelligence through Reinforcement Learning

Next-generation AI compels us to realize that machines do indeed think. Although machines do not think like us, their thought process has proven its efficiency in many areas. In the past, the belief was that AI would reproduce human thinking processes. Only neuromorphic computing (see Chapter 18, Neuromorphic Computing), remains set on this goal. Most AI has now gone beyond the way humans think, as we will see in this chapter.

The Markov decision process (MDP), a reinforcement learning (RL) algorithm, perfectly illustrates how machines have become intelligent in their own unique way. Humans build their decision process on experience. MDPs are memoryless. Humans use logic and reasoning to think problems through. MDPs apply random decisions 100% of the time. Humans think in words, labeling everything they perceive. MDPs have an unsupervised approach that uses no labels or training data. MDPs boost the machine thought process of self-driving cars (SDCs), translation tools, scheduling software, and more. This memoryless, random, and unlabeled machine thought process marks a historical change in the way a former human problem was solved.

With this realization comes a yet more mind-blowing fact. AI algorithms and hybrid solutions built on IoT, for example, have begun to surpass humans in strategic areas. Although AI cannot replace humans in every field, AI combined with classical automation now occupies key domains: banking, marketing, supply chain management, scheduling, and many other critical areas.

As you will see, starting with this chapter, you can occupy a central role in this new world as an adaptive thinker. You can design AI solutions and implement them. There is no time to waste. In this chapter, we are going to dive quickly and directly into reinforcement learning through the MDP.

Today, AI is essentially mathematics translated into source code, which makes it difficult to learn for traditional developers. However, we will tackle this approach pragmatically.

The goal here is not to take the easy route. We're striving to break complexity into understandable parts and confront them with reality. You are going to find out right from the outset how to apply an adaptive thinker's process that will lead you from an idea to a solution in reinforcement learning, and right into the center of gravity of the next generation of AI.