All of the basic search agents are discussed using the Berkeley AI materials of Pacman interface.

Reinforcement Learning (RL) is one of the model free machine learning algorithms where the agent learns its behaviours from the environment by actually interacting with it. This is better than the offline planner because the agent actually interacts with the environment to learn its behaviours because it is almost impossible to simulate a real world in a computer. By using the reinforcement learning, the agent learns those extra features which can only be learned in an real world environment hence giving it a learning capability like living organisms because in a real world there are certain parameters which cannot be simulated by a computer. Since the reinforcement learning agent gets its feedback from the environment, it allows the agent to automatically determine its behaviours that are considered ideal within a specified context. Reinforcement learning is deemed important in the field of artificial intelligence as it starts to make breakthrough and benchmarks in various industrial applications. Previously we have analysed the pacman game where the pacman agent is a reflex agent, here, we are trying to make the pacman agent more smarter by applying RL techniques, i.e, Q-learning successfully.