Frequently asked questions
Q: What is the difference between supervised learning and reinforcement learning?
A: In the case of supervised learning algorithms, the model is trained based on historical data which describes the trend for the data historically and establishes a correlation between the event data and resultant output. In that case, the supervised learning model is a curve fitting exercise that maps the data points (independent variables) to a set of output variables (dependent variables). Availability of the historical data is essential for supervised learning. In case of reinforcement learning, the agent is modeled based on the rewards it receives based on the action(s) it takes within the context of the environment in which it is operating. There is no historical data available to the agent to train itself. However, a hybrid approach often works great where the agent is aware of the historical trends as well as applies exploration and exploitation strategies in order to maximize...
