Understanding Reinforced and Semi supervised Machine learning

Article 1:Iintroduction to Artificial intelligence and Machine learning

Article 2:Introduction To Machine Learning Algorithm
article 3:Understanding Supervised and unsupervised machine learning

Understanding Reinforced and Semi supervised Machine learning

As we already discussed the first two types of Machine learning Algorithm its time to understand next two and final types of machine learning algorithm which are:

*Semi supervised Machine learning
*Reinforced Machine Learning

Review of Supervised and Unsupervised Machine learning.

Supervised Machine learning:

In supervised machine learning ,system is trained using training data or also labeled data or also classified example modal .algorithm uses these training data to take decision and show results ,this type of algorithm is very time consuming and expensive but more accurate.

Unsupervised machine learning:

In unsupervised machine learning ,system is trained without any labeled data or also classified data or also example modal. These type of algorithm is less time consuming and expensive but less accurate.

What is semi supervised machine learning?

Due to use of a lot of labeled data ,supervised learning is very time consuming and expensive,in some experiment data scientists used some labeled data with some unlabeled data and founded that it increases the accuracy of result and it was easy to implement than supervised learning. And more accurate than unsupervised learning.
         Hence they devloped a new algorithm which falls between supervised and unsupervised learning aka,semi supervised learning.
       Semi supervised learning involves human agent to render data and modals .
       

As we now know what is semi supervised learning we can now try to understand  About Reinforced machine learning..

What is Reinforced machine learning?

Reinforced learning is a technique to train a program using trial an error method or more practical way as human learns from environment using their actions.

For example if a kid does something good gets a choclate as a reward and  for doing something bad gets scolded by their parents.
      Reinforced learning is also based on same principal if system does something valuable gets reward points by agents but when it does something value less ,reward point is reduced .
System used these past experience to learn and be accurate .
        So simply this technique is more action based and trained under live environment , this topic is very eye catching for research among data scientists and programmers due to its possibility in future to create a very advanced independent system.
       In next we will discuss more about these technique and terms related to Artificial intelligence .

If you want to join our official Telegram group, CLICK HERE

Written by- Aakash deep gupta