Reinforcement learning model for virtual pets

Abstract: Virtual pets are developed with the ability to learn from their environment and interact with the user, which is used for Reinforcement Learning and interaction local or independent. Intelligent agents are used in a virtual environment and receive feedback; we simulate many daily activities of a dog which is essential for the interaction with the user. The use of learning in pets gives a real dimension artificial life and allows the system to evolve intelligently. They give a satisfactory learning in mathematical linguistics considering it as a process. The sequence is repeated at different intervals of time, therefore it is a recursive function, and finally obtained the generalization error, through the concept of distance.

Keywords: Learning, computing, artificial intelligence, pet.