Learning to Avoid Risky Actions

When a reinforcement learning agent executes actions that can cause frequently damages to itself, it can learn, by using Q-learning, that these actions must not be executed again. However, there are other actions that do not cause damage frecuently, only once in a while: risky actions, such as parachuting. These actions may imply a big punishment to the agent and, depending on its personality, it would be better to avoid. Nevertheless, using the standard Q-learning algorithm the agent is not able to learn to avoid them, since the result of these actions can be positive in average. In this paper, an additional mechanism to Q-learning, inspired by the emotion of fear, is introduced in order to deal with those risky actions by considering the worst results of them. Moreover, there is a daring factor for adjusting the consideration of the risk. This mechanism is implemented on an autonomous agent living in a virtual environment. The results present the performance of the agent with different daring degrees.