Goal-based agents

- Let's take a look at goal-based AI agents, which improve on model-based agents because information about the goal is included. These agents are also called planning or rule-based agents because they use their knowledge and a set of rules to plan and choose the best strategy to achieve a specific goal. An example would be a chess playing AI whose goal is winning the game. Another difference is that goal-based agents can be proactive considering a long sequence of potential actions before deciding whether it can achieve the goal or not. The result is that goal-based agents ensure that every option is considered before an action is taken, which creates optimum outcomes while ensuring cost efficiency and better resource management business areas that would benefit from goal-based agents are those that have goals that need to be managed and maintained routinely. For example, maintaining a customer satisfaction score requires identifying when a score decreases, analyzing the source of the problem, and then developing an action plan of changes to make. Other areas ripe for exploration includes supply chain management, talent management, and product or service innovation. The disadvantage of goal-based agents is that they have limited scope because they focus on goals defined by the environment. If new or other factors come into play that are not already included in the defined objective, the goal-based agent may not take them into consideration when making decisions. They can also be time consuming in resource intensive to train and fine tune, and you may not know if it will work until a healthy investment is already made. And lastly, goal-based agents are hardcoded to achieve specific objectives. This means they have problems solving unique and unfamiliar problems since they follow pre-program rules. Now, one big advantage of goal-based agents is that they can be combined with other AI techniques to create more advanced agents. Because of their ability to adapt to changes in their environment, they can learn and modify its strategies. This makes them ideal building blocks, combining them with other AI techniques to create more advanced agents. So let's see how goal-based agents impact the next type utility based agents.