Artificial Intelligence

Any technique which enables computer to mimic human behavior

Machine Learning

Subset of AI techniques which use statistical methods to enable machines to improve with experiences

Deep Learning

Subset of ML which make the computation of multi-layer neural networks feasible

Type of Machine Learning Techniques

• Supervised
  ” Optimize performance criteria using experience”
  
• Unsupervised
  ” Find features for categorization”
  ” Taken place in real time”
  
• Reinforcement
  ” To maximize some notion of cumulative reward”

Comparison Supervised versus Unsupervised

Parameters	Supervised	Unsupervised
Process	Input & Output will be given	Only Input data will be given
Input Data	Labeled	Unlabeled
Algorithms	SVM (Support Vector Machine), Neural Network, Linear and Logistics Regression, Random Forest, Classification Trees	Clustering Algorithm, K-Means, Hierarchical clustering
Computational Complexity	Simpler	Complex
Use of Data	Training Data (Link Input with Output)	Doesn’t use Output Data
Accuracy of Results	Highly accurate	Less accurate
Real Time Learning	Offline	Real time
Number of Classes	Known	Unknown
Main Drawback	Big Data can be a real challenge	Cannot get precise information