Ingredients of AI

Artificial Intelligence (AI)

Possibility of inanimate objects, such as machines, robots, systems, etc., with computer ability to carry out intelligent activities similar to those performed by people.

Examples of AI

• Vision (Video Analytics & Image Processing)
• Hearing (Speech to Text Applications)
• Response to Stimuli (Inputs)
• Cognition

Click here to learn Artificial Intelligence in Bangalore, Artificial Intelligence in Hyderabad

Data Science

Data science is a discipline of study that focuses on data in order to produce actionable business insights.

Topics of Data Science includes

• Statistical Analysis
• Hypothesis Testing
• Data Visualization
• Regression Analysis
• Classification Techniques
• Black Box Techniques
• Text Mining
• Natural Language Processing
• Time Series Analysis, etc.

Click here to learn Data Science in Hyderabad, Data Science in Bangalore

Data Mining

Similar to coal mine, data mining yields coal and, with chance, valuable stones like diamonds. In data mining, we extract insights from the data, and insights that resemble diamonds are very beneficial to enterprises.

Machine Learning

Machine learning is the process of discovering patterns in history or previous data and applying that knowledge to previously unknown future data in order to accomplish a certain goal.

• Machine Learning Supervised Learning - Both Inputs and Output are known in the historical data
• Machine Learning Unsupervised Learning - Only Inputs are known in the historical Data & Output variable is not known or assumed as not known

Click here to learn Machine Learning in Hyderabad, Machine Learning in Bangalore

Deep Learning

In a particular area of machine learning called deep learning, the underlying patterns in data are automatically retrieved.

Some of the Deep Learning Architecture

• Artificial Neural Networks / Multi-Layered Perceptron
• Convolutional Neural Network
• Recurrent Neural Network
• Deep Belief Network
• Long Short Term Memory (LSTM)
• Gated Recurrent Units (GRUs)
• Mask R-CNN
• Autoencoders
• Generative Adversarial Network (GAN)
• Boltzmann Machine
• Deep Q-Networks
• Q Learning etc.

Reinforcement Learning

Games, robotics, investment banking, trading, and other applications significantly employ reinforcement learning, a specific branch of deep learning.

Reinforcement learning is reward-based learning that uses interaction with the environment to solve sequential decision-making issues.

The 5 key elements of Reinforcement Learning

• Agent: Agent is a learning component that makes a decision on actions to maximize the reward.
• Policy: Policy defines the behaviour of the agent from states to actions.
• Environment: Environment is the physical world where agents perform actions.
• Value Function: Value Function defines the cumulative future reward.
• Reward Function: Reward Function defines the problem and maps it to a numerical reward.

An optional component that forecasts the behaviour of the environment is the Model of the Environment.

Stages of Analytics

• Descriptive Analytics - Answers questions on what happened in the past and present.

  Example: Number of Covid-19 cases to date across various countries.

• Diagnostic Analytics - Answers questions on why something happened.

  Example: Why are the Covid-19 cases increasing?

• Predictive Analytics - Answers questions on what might happen in the future.

  Example: What will be the number of Covid-19 cases for the next month?

• Prescriptive Analytics - Provides remedies and solutions for what might happen in the future.

  Example: What should be done to avoid the spread of Covid-19 cases, which might increase in the next month?