ARTIFICIAL INTELLIGENCE LEARNING PROCESS

ARTIFICIAL INTELLIGENCE LEARNING PROCESS

WHAT IS LEARNING IN ARTIFICIAL INTELLIGENCE ?

The main property of Artificial Neural Network is to learn.

Learning is a process by means of which a Neural Network adopts itself to simulate by making proper parameter adjustments.

Types of learning are :

Supervised Learning

Unsupervised learning

Reinforcement Learning

SUPERVISED LEARNING

In Artificial Intelligence following the supervised learning where each input requires a corresponding target vectors which represents the desired output.

The input vector along with the target vector is called training pair or training data.

During training the input vector is presented to the network which results each an output vector.

The output vector is known as actual format of output vector.

Then this output vector is compared with the desired output vector.

If there exist a difference between the output vectors and desired vector, then an error signal will be generated by the network.

This error signal is used for adjustment of weight with the actual output matches with te desired output.

for this reason in this type of training a supervisor is required for error minimization. Hence the network trained by this method is called as SUPERVISED TRAINING PROCESS.

2. Unsupervised Learning

Here the learning process will be perform without the help of a trainer.

In this type of Artificial Neural Network the input vectors of similar type are grouped without the use of training data to specify how a member of each group look or to which group a number returns.

In this training process the network receives the input pattern and recognizes this pattern to form Clusters.

When a new input pattern is enter to the network then to findout the actual output the network match all the input to the available clusters. This feature is known as Self Organizing Feature or Cluster Technique.

3. Reinforcement Learning

The reinforcement learning process is very much similar to the supervised learning. In this case of learning, the network might be told that its actual output is only 50% correct.

So this information is called Critic Information to the network.

This type of learning which is based upon Critic information to the network is called as Reinforcement Learning.

After getting on reinforcement signal or feedback signal will be generated by the network.

Thank You,

With Regards

Rakesh Kumar Sen

Comments