Artificial Intelligence | Definition | How AI works? | Neural network | Machine Learning | Self Driving Cars

 Artificial Intelligence

    Artificial Intelligence is the intelligence of the machines that make them able to do things that a human brain does. With the help of Artificial intelligence, machines are able to make complex decisions based on the predictions made with the help of Machine learning or Deep Learning. 

   Machine Learning and DeepLearning plays a very important role in the execution of artificial intelligence as they help machines to learn themselves without being programmed again and again. 

How AI works?

    Artificial Intelligence means computer has the ability to work on its own. For this computer needs to make certain decisions and predictions on their own. But computer are non living machines and to make decisions and predictions on its own it needs a brain to process the information. Here machine learning comes into the picture. Its makes the computer learn from the environment. It uses algorithms to do such work.

   Algorithms are the programs that are feed by the programmers in the beginning.Then these algorithms are use the neural network to get the raw data to process. Here neural network work as a brain of the AI where all the data is stored in the form of neurons. The first neuron created was called Perceptron. It stores the data in the form of binary. 

     These neural network gets activated by an algorithm which uses the raw data in it. Then based on the raw data the computer makes decisions and predictions and find the high accuracy output for the given input. 

Some real world Applications

1. Marketing: Companies use AI to advertise their products and make them look more attractive to its customers.
2.  Banking: Banks use AI for customer support, detect anomalies and credit card frauds.
3.  Agriculture: AI helps the farmers to get more from a land using the rersources more sustainably.
4. Finance: In this field AI is used to predict the future patterns in market. This helps the people to trade in different sectors
5.  Healthcare: AI has achieved more accuracy and precision in going what a good medical facility does. It has increases the success rate by doing complex operations with high accuracy and precision.
6. Gaming: AI has been get into the market first in gaming industry. It has achieved so much that we get an extra ordinary experience in gaming.
7. Space Exploration: AI and machine learning helps space exploration to a higher extent. It helps to analyze and process huge amount of data gathered during the exploration.

8.     Autonomous Vehicles: AI and advance deep learning has made it easier for the vehicles to operate on their own and we are able to create self driving cars.

9.     Chatbots: AI is used in the chatbots like virtual assistant which talk to us like that they are just like us.

10.     Artificial Creativity: AI has made many creative innovations in the digital market from face unlock to the voice recognition.

Machine Learning | Definition | Artificial Intelligence | Deep Learning | Applications | Algorithm | Supervised Learning | Unsupervised Learning | Reinforcement Learning

Machine Learning

   Machine Learning is a study of computer algorithms in which machines are trained to do a particular task. It is the field of Artificial Intelligence where it uses the data (training data), which is a stored in it to make predictions or decisions for some input. It gives the machines ability to learn and improve from its experiences without being programmed. 

   But it can be used for basic operations which require small data, for complex operations we use deep learning which doesn’t require data. It has its useful  application in computer vision and email filtering.

Three Ways of Machine Learning 

Supervised Machine Learning:

In this learning, the data present (called as predictor variables/features) in the machine are labeled with respect to the output data (target variable). This predictor variable make predictions according to the input data and goes on until it reaches the higher accuracy result with respect to the target variable.

Algorithms:

1. Linear regression

2. Logistic regression

3. Support Vector Machines

4. Naive Bayes

5. K-nearest neighbour algorithm

6. Random Forest Algorithm

Applications:

1. Bioinformatics

2. Quantitative structure

3. Database marketing

4. Handwriting recognition

5. Information retrieval

6. Learning to rank

7. Information extraction

8. Object recognition in computer vision

9. Optical character recognition

10. Spam detection

11. Pattern recognition

Unsupervised Machine Learning:

In this learning, the data present data present in the machine is unlabeled, i.e. it does not have the target variable to make predictions, so it uses the sequence of the input data to make certain predictions. For eg. The customers are categorized on the basis of the amount of products bought from company without knowing in advance the category.

Algorithm:

1. K-means Algorithm

2. Apriori Algorithm

3. Expectation–maximization algorithm (EM)

4. Principal Component Analysis (PCA)

 Application :

1. Human Behaviour Analysis

2. Social Network Analysis to define groups of friends.

3. Market Segmentation of companies by location, industry, vertical.

4. Organizing computing clusters based on similar event patterns and processes.

Reinforcement Machine Learning :

In this learning, the machine is programmed to perform some actions in the environment and in return it gets reward or punishment. On the basis of that reward or punishment, it programs itself to improve and perform further actions. This process of learning and improving goes on.

Algorithm:

1. Q-Learning Algorithm

2. State–action–reward–state–action Algorithm (SARSA)

3. Deep Q Network Algorithm (DQN)

4. Deep Deterministic Policy Gradient Algorithm (DDPG)

Application:

1. Resources management in computer clusters

2. Traffic Light Control

3. Robotics

4. Web System Configuration

5. Personalized Recommendations

6.Deep Learning

Deep Learning | What is Deep Learning | Neural Network | Machine Learning | Artificial Intelligence | Books

What is Deep Learning?

   

       Deep Learning is a broader concept of machine learning. Deep learning is basically learning from the real life world with the help  of neural network and giving appropriate output at present time.

     Neural network can be called as the brain where all the data is present. Neural network is connection between millions of raw data that is been used for deep learning process. It is very similar to the machine learning in some functions as both are used to learn from the real life world.

     One main advantage in deep learning is that it does not require any specific information for learning , in-fact it generates the features for the object and give the appropriate output. Whereas in machine learning, the features are inserted for a specific object. That is the reason that machine learning are not used for complex problems, instead deep learning is used.

Why we need Deep Learning?

     In this fast growing world, we have a lots of objects which contain a lots of data which does complex jobs. Here complex jobs are referred to the work which requires a huge amount of data. This complex data needs to be analysed and interpreted for the desired output. 

     But the machine learning as uses small objects whose features are already present. So to overcome this problem deep learning is used. 

     It can handle a huge amount of data and does not require the features to be present. Deep learning is so capable that it generates the features of the object and give the appropriate output. So it is the one of the important features of Artificial Intelligence.   

Must read Books for Deep Learning      

       

 Deep Learning (Adaptive Computation and Machine Learning series)

    Deep Learning with Python

Hands-On Deep Learning Algorithms with Python: Master deep learning algorithms with extensive math by implementing them using TensorFlow 

Neural Networks and Deep Learning: A Textbook

                      

Useful Applications of Deep Learning

• Self driving cars: Deep learning algorithms are used in identifying objects so that car can make appropriate decisions while driving.
• Language Processing : Deep learning helps in language translation in multiple languages.
• Medical : Deep learning helps doctors to find tumor for cancer like cells and its spread in body. 
• Uber : It also helps in analyzing and displaying the suitable route and required time for the destination. on the basis of previous data from same destination.