(Reinforcement Learning): In reinforcement learning, the reward is the feedback provided to an agent when a specific action is taken at a particular state in the environment. Using the concept of “trial and error” for reinforcement learning, the rewards are the “errors”. For example, if an agent is trying to find a solution to a maze, the reward would be a positive point if it makes progress towards the solution in a maze, and a negative point if it encounters an obstacle or backtracks on an existing path.
Login to register for events. Don’t have an account? Just register for an event and an account will be created for you!