when we are dealing with machine learning we come to know a area of machine learning known as reinforcement learning which is very interesting and in this article on Thearensic we will talk about what is reinforcement learning?and some examples of Reinforcement learning
what is reinforcement learning?
if you have some knowledge about machine learning you already know about the agent and the environment.so the agent navigate through the environment and learn from the feedback’s.basically the agent performs some actions and from the action the agent got some reward mainly +1 and -1.
At time t, the agent perceives state S and reward R from the environment the reward can be -1 or +1 it depends on the state, and takes an action A, which will affect the state of the environment and after taking the action the agent got into new state. The environment, at time t+1, generates another state and reward. The agent evaluates the goodness of a state using value function, and makes decisions according to a policy.and this process repeats and agent learns the whole environment.
after performing the action two things happens
- the agent is got into new stage
- the agent got some reward.
while changing the states the agent learns everything on that environment and then become the master of that environment.
we can compare agent from our daily life like for us our environment is our home and the agent is us.we perform different actions and also leads to some reward and after performing that action we got some experience from that action and we learned something from that action similar to that agent who learns from the environment.
conclusion: reinforcement learning is the branch of machine learning which deals with the learning of agent by doing certain actions for which the reward is maximum.reinforcement learning has an agent and a environment in which the agent learns by performing some actions and also getting the reward for that particular action.
Types of Reinforcement
there are two types of reinforcement learning namely Positive Reinforcement and negative Reinforcement learning .
when an incident happens due to a specific conduct, improves the behavior’s power and frequency. which has a positive impact on behavior. this type of Reinforcement is known as positive Reinforcement.
the Negative Reinforcement learning is described as a strengthening of behavior because a negative condition is obstructed or avoided or the condition which give the reward of -1 is avoided.
Difference between Reinforcement learning and Supervised learning
as the name suggest supervised that means somebody or some data tell the agent what is actually correct for example your school teacher they known what is the right answer of a question and they can easily point out tell you about this.
Reinforcement learning is the process of leaning through the feedback of Agent on an environment and this area of Machine learning can be apply on the software as well as hardware. this type of learning learns from the feedback for the action that the agent has performed.
Reinforcement learning in marketing
since it is the world of marketing where each and every company is focusing on their marketing Reinforcement learning come to help them big Ad companies like google to display personalize Ads to the users so that maximum conversion can happen.
The RL helps the companies to spend minimum in their promotions and gain as much as profit by creating a relevant Eco system of ads for the users we all known how google works on their ads to show the relevant Ads so that the companies and individuals can progress on their marketing.
thanks for visiting guys i”ll catch you on the next one till then,
you can also read these: