Giving Agents Intelligence

While making an agent intelligent is not a simple task, there are a variety of methods that can be used to endow an agent with reasonable decision-making capabilities. Some of these methods are discussed within this book (see Figure 11.2).

Figure 11.2: Providing agents with decision-making capabilities.

The creation of game AI characters is discussed in Chapter 5 using the backpropagation neural network model. One key characteristic of neural networks is that they're not only capable of generalizing (making good decisions for unforeseen circumstances); they also can adaptively modify their behavior given changes in their environment.

The use of forward-chaining, rules-based systems is discussed in Chapter 8. The rules-based system permits an agent to reason about its environment given a set of rules and facts that describe its environment. An important characteristic of rules-based systems is that they can describe their line of reasoning (what facts lead to a particular decision).

Clustering algorithms (such as adaptive resonance theory, described in Chapter 3) are another useful technique for agent intelligence. These algorithms permit agents to identify relationships within their environment. This can help an agent learn patterns without any instruction ( unsupervised learning).

Numerous other methods can be used, some of which were described in prior chapters. Fuzzy logic is also useful as an element for intelligent agents (described in Chapter 9), as is simulated annealing (described in Chapter 2) for constraint satisfaction.